Fluid Responsiveness Versus Fluid Tolerance

A fluid infusion can increase cardiac output and worsen congestion at the same time. Before giving more fluid, establish whether the patient needs better perfusion. Then assess whether additional preload will increase blood flow and whether the patient can tolerate the resulting changes in pressure and tissue fluid.

20.1 What Fluid Responsiveness Tells Us

Fluid responsiveness means that stroke volume or cardiac output increases when cardiac preload increases. It describes how the circulation responds at that moment. It does not establish that the patient is dehydrated or that fluid is the best treatment.

Preload refers to the stretch of myocardial fibers before contraction. When more blood returns to a preload-responsive ventricle, its end-diastolic volume increases. The additional stretch allows the ventricle to eject more blood through the Frank–Starling mechanism.

This response becomes smaller as the ventricle approaches the flatter part of its function curve. Additional filling then produces little increase in stroke volume. Filling pressure may continue to rise even though forward flow barely changes.

Both ventricles must handle the additional blood. Blood entering the right ventricle must pass through the pulmonary circulation before reaching the left ventricle. A failing right ventricle may not deliver that blood effectively. A stiff or failing left ventricle may receive it only at a much higher filling pressure.

We express the measured response as a change from baseline:

Suppose the heart pumps 60 mL with each beat before a fluid challenge. After the challenge, it pumps 69 mL with each beat. The increase is 9 mL.

To express that increase as a percentage:

9 ÷ 60 × 100 = 15%

The heart is therefore pumping 15% more blood with each beat after fluid administration.

We do not classify every small increase as fluid responsiveness. A tiny change may reflect normal variation between heartbeats or measurement error. Many fluid-challenge studies therefore require an increase in stroke volume or cardiac output of about 10–15% before calling the response positive. 1

For a method that uses a 15% cutoff, an increase from 60 to 69 mL counts as a positive response. An increase from 60 to 61 mL does not.

A positive response tells us that fluid increased the amount of blood pumped per beat. We must then decide whether that increase improved perfusion and whether the fluid caused harm.

Responsiveness changes as treatment changes the circulation. An infusion may move the ventricles onto a flatter part of their function curves. A vasopressor may recruit blood from venous reservoirs. Changes in ventricular function, rhythm, or ventilation can also alter the response. A positive test earlier in the day does not establish that the patient remains responsive now.

20.2 Establish the Need for Fluid Before Testing

A patient can be fluid responsive without needing fluid. Preload reserve simply means that the heart can eject more blood if filling increases. There is no reason to eliminate that reserve when perfusion is already adequate.

Start by identifying the problem that an infusion is intended to correct. A patient who has lost circulating volume may have insufficient venous return and stroke volume. In that setting, replacing the loss can improve cardiac output and tissue perfusion. The same treatment may offer little benefit when another mechanism limits the circulation.

Consider hypotension caused mainly by vasodilation. Arterial pressure may remain low despite an adequate cardiac output because vascular resistance is low. More fluid may increase filling pressures without adequately correcting the loss of vascular tone.

Oliguria also requires an explanation. It may reflect reduced renal perfusion, but venous congestion or intrinsic kidney injury can produce the same finding. Giving fluid solely because urine output is low can worsen congestion when the kidney is already struggling to drain blood.

An elevated lactate is similarly incomplete evidence. It can accompany inadequate oxygen delivery, but changes in lactate production and clearance can also raise its concentration. The clinician must connect the laboratory result to the rest of the examination.

The history of losses, peripheral perfusion, cardiac findings, and the response to previous treatment help establish whether insufficient preload is contributing. Responsiveness testing is most useful when that explanation is plausible and the expected flow response remains uncertain.

Urgency changes the sequence. Major ongoing hemorrhage requires bleeding control and appropriate replacement. Testing should not delay these actions. A stable patient with an uncertain cause of low output allows more time to clarify the mechanism.

Choose the replacement according to what has been lost. Crystalloid can expand extracellular volume but does not replace the oxygen-carrying capacity of red cells. Electrolyte-free water distributes differently from a sodium-containing resuscitation fluid. Predicting a rise in cardiac output does not determine which fluid the patient needs.

20.3 How Fluid Can Increase Flow and Still Cause Harm

Fluid tolerance describes the ability to receive additional fluid without clinically important harm. Assessing it requires more than predicting cardiac output. We must also consider how much filling pressures will rise, where fluid may accumulate, and whether the patient can excrete it.

Ventricular compliance determines how much pressure rises as the chamber fills. A compliant ventricle accommodates additional blood with a relatively small pressure increase. In a stiff ventricle, even a modest increase in volume can produce a large rise in filling pressure.

When left ventricular filling pressure rises, pressure can rise in the left atrium and pulmonary veins. Higher pulmonary microvascular pressure favors movement of fluid into the lung interstitium. If filtration exceeds lymphatic removal, lung water increases. Further accumulation can flood alveoli and impair gas exchange.

The right side creates a different problem. If the right ventricle cannot eject the additional blood, right atrial and systemic venous pressures may rise. This reduces the pressure difference available for flow through some organs and increases the tendency for fluid to leave their microvessels.

The ventricles also influence each other. They share a septum and lie within the pericardium. When a right ventricle becomes markedly distended, it can shift the septum toward the left ventricle and limit left ventricular filling. Adding volume may then worsen the filling problem it was intended to correct.

Barrier integrity determines how readily increased vascular pressure produces tissue edema. An injured pulmonary or systemic microvascular barrier allows more fluid, and sometimes more protein, to cross into tissue. Existing edema and impaired lymphatic drainage make further accumulation harder to clear.

Renal function determines how long the added sodium and water remain in the body. The immediate increase in cardiac output may disappear as fluid redistributes. If excretion is impaired, that fluid can remain in the interstitium long after its circulatory benefit has faded.

These effects explain why tolerance differs between organs. Si and colleagues studied changes in cardiac index, venous congestion, and lung water during fluid challenges. Venous congestion worsened mainly in nonresponders. In patients with acute respiratory distress syndrome, however, lung water could increase despite a rise in cardiac index. A favorable flow response therefore did not establish pulmonary tolerance. 2

No single measurement has been validated to classify tolerance across all ICU patients. Describe the specific concern instead: rising left-sided filling pressure, a distending right ventricle, worsening venous congestion, increasing lung water, or impaired excretion. This tells the reader what may go wrong and why.

20.4 Test Preload Responsiveness with a Reversible Maneuver

A reversible preload maneuver briefly increases cardiac filling while stroke volume or cardiac output is measured. It tests the proposed treatment mechanism without leaving an additional infusion in the body.

Passive leg raising uses a controlled postural change to move blood from venous reservoirs toward the thorax. This increases venous return and tests whether the ventricles can turn the additional filling into greater forward flow. Returning the patient to the starting position removes the preload stimulus.

The response must be measured while the effect is present. A monitor that averages over too long an interval may miss the peak change. A return toward baseline after repositioning also helps confirm that the observed change was related to the maneuver.

Monnet and colleagues showed that changes in aortic blood flow during passive leg raising predicted the response to volume expansion. Their study also showed why the maneuver can be useful when arrhythmia or spontaneous breathing makes respiratory pulse-pressure variation unreliable. 3

Measure flow rather than relying only on arterial pressure. Pressure depends on cardiac output, vascular resistance, and arterial properties. Stroke volume can rise without a proportionate pressure increase. Conversely, a pressure rise does not establish that stroke volume increased.

An end-expiratory occlusion tests preload through a different mechanism. During positive-pressure ventilation, each inflation can impede systemic venous return. A brief hold at end-expiration interrupts those cyclic inflations. Venous return can increase, and a preload-responsive circulation may increase cardiac output. Monnet and colleagues demonstrated the predictive value of this response in selected ventilated ICU patients. 4

The patient must tolerate the maneuver, and the monitor must detect a small change promptly. Postural restrictions, an ineffective transfer of venous blood, unstable conditions, or inadequate measurement can prevent a valid assessment. Failure to produce or measure the intended stimulus makes the test uninterpretable; it does not establish nonresponsiveness.

Use the threshold validated for the particular maneuver and measurement method. A cutoff for a fluid challenge may differ from one used for passive leg raising or an end-expiratory occlusion. When the measured change lies close to that cutoff, measurement uncertainty makes the classification less secure.

A positive result still leaves the tolerance question unanswered. Briefly recruiting the patient's own blood does not reproduce the later effects of retaining additional sodium and water.

20.5 When Respiratory Variation Becomes Misleading

Positive-pressure ventilation creates a repeated cardiovascular disturbance. During inflation, intrathoracic pressure rises and systemic venous return often falls. Lung inflation can also increase right ventricular afterload. Both effects can reduce right ventricular stroke volume during part of the respiratory cycle.

After blood passes through the pulmonary circulation, the change in right ventricular output alters left ventricular filling. Left ventricular stroke volume then varies across the respiratory cycle. When arterial conditions are reasonably stable, arterial pulse pressure also varies with the ejected volume.

Pulse-pressure variation and stroke-volume variation use these changes to predict preload responsiveness. Michard and colleagues demonstrated the value of pulse-pressure variation in selected mechanically ventilated septic patients. The result depends on the conditions under which the respiratory disturbance is generated. 5

Spontaneous effort changes intrathoracic pressure in an irregular and sometimes opposite direction from passive inflation. Arrhythmia changes stroke volume through variable filling time. Either can create beat-to-beat variation that no longer reflects a consistent ventilator-induced preload test.

Low tidal volume may create too small a disturbance to reveal preload dependence. Low respiratory-system compliance can also change how airway pressure is transmitted to the pleural space and pulmonary vessels. A low variation under these conditions may fail to identify a responsive patient.

A high variation can be misleading for another reason. If inspiration markedly increases right ventricular afterload, right ventricular output may fall even when added volume will not help. Mahjoub and colleagues demonstrated false-positive pulse-pressure variation associated with right ventricular dysfunction. 6

This is especially important when the right ventricle is dilated and systemic venous pressure is high. Additional fluid may further distend the ventricle without correcting the pulmonary vascular resistance or contractile failure that limits ejection.

When the assumptions are not met, select another assessment method. Changing the numerical cutoff does not correct an invalid physiological test. A valid response must be linked to the stimulus, timing, and flow measurement.

20.6 Give a Fluid Challenge with a Defined Aim and Stopping Limits

A fluid challenge gives a specified fluid volume over a specified interval and measures the response. It directly tests whether adding fluid increases flow. It also exposes the patient to the fluid's immediate and delayed effects.

Define the intended benefit before administration. For example, an increase in stroke volume may be the immediate measurement, while improved peripheral perfusion is the clinical aim. These should be connected explicitly so that an isolated monitor change is not mistaken for successful treatment.

Define the limits as well. A marked rise in filling pressure, worsening respiratory status, or failure to increase flow may indicate that further administration is unlikely to help. The relevant limits depend on the organ most at risk.

The volume and infusion rate must create a measurable preload change. If the exposure is too small for the monitor to detect, an apparent negative result may reflect an inadequate test. A larger or faster exposure may produce a clearer signal but also a greater rise in pressure. No single challenge volume is appropriate for every patient and measurement method.

Keep measurement conditions comparable whenever possible. A change in vasopressor dose can alter vascular resistance and venous return. Changes in rhythm, ventilation, or sedation can independently alter stroke volume. When these interventions occur together, their effects cannot all be attributed to the fluid.

The FENICE cohort documented substantial variation in how clinicians performed fluid challenges, assessed the response, and used stopping limits. The study describes practice rather than proving one optimal protocol. It illustrates why giving a bolus without defining what will be measured provides less useful information. 7

Compare the pressure response with the flow response. If filling pressure rises substantially while stroke volume barely changes, the infusion has failed to produce the intended increase in ejection. The higher upstream pressure can increase venous congestion and filtration into tissue.

Reassess before repeating the challenge. The first infusion may have used up the available preload reserve or worsened congestion. A successful first bolus does not establish that the next one will produce the same balance of benefit and harm.

20.7 Check Whether the Benefit Lasts

The purpose of increasing cardiac output is usually to improve tissue perfusion and oxygen delivery. Cardiac output is only one part of oxygen transport:

With cardiac output in L/min and arterial oxygen content in mL of oxygen per dL of blood, oxygen delivery is expressed in mL/min. The factor of 10 converts liters to deciliters.

Consider an illustrative cardiac output of 4.0 L/min and arterial oxygen content of 18 mL/dL. Oxygen delivery is 720 mL/min. If output increases to 4.6 L/min while oxygen content falls to 17 mL/dL, oxygen delivery becomes 782 mL/min.

Cardiac output increased by 15%, but oxygen delivery increased by about 8.6%. The lower oxygen content offsets part of the flow gain. Hemodilution is one possible explanation, although the assumed numbers do not predict the dilution produced by any particular fluid dose.

More global oxygen delivery also does not ensure that every tissue receives or uses more oxygen. Regional distribution can remain uneven. Microvascular obstruction, increased diffusion distance, impaired extraction, or disturbed cellular metabolism can limit the benefit. Follow the perfusion problem that prompted treatment rather than cardiac output alone.

The response may also fade. Fluid moves out of the plasma compartment, some is excreted, and vascular tone changes. The heart may therefore receive less additional preload at a later measurement than it did immediately after the infusion.

In the FCREV study, Roger and colleagues followed the response to fluid challenges in sepsis and identified both persistent and transient responders. The study shows why the time of measurement is part of the result. An immediate rise in stroke volume does not establish that the increase will persist. 1

A transient improvement can still be useful during resuscitation. The problem arises when repeated boluses replace a fading effect while the underlying cause of shock remains untreated. Fluid may then accumulate in tissue without providing sustained circulatory improvement.

20.8 Assess the Heart, Lungs, and Venous Circulation

Fluid tolerance cannot be read from one organ or one number. Examine how additional filling is likely to affect the heart, where pressure may be transmitted, and how fluid is being cleared.

Cardiac ultrasound can identify a dilated or poorly functioning right ventricle, major valvular disease, or impaired left ventricular filling. These findings help explain why filling pressures may rise sharply with additional volume. Chamber size alone does not predict how much more blood the heart can accept.

Interpret filling pressure in relation to surrounding pressure. The pressure that distends a cardiac chamber is its transmural pressure: the pressure inside minus the pressure outside. Positive intrathoracic pressure can raise the measured intracavitary pressure without an identical increase in chamber distension. The ventilator and cardiac findings therefore need to be interpreted together.

Assess the lungs with the available clinical and imaging information. Increasing oxygen requirements, worsening respiratory mechanics, and new ultrasound abnormalities may indicate that fluid is accumulating. B-lines arise when changes near the pleural surface produce vertical ultrasound artifacts. They support an alveolar–interstitial abnormality but do not identify its cause. Hydrostatic edema, inflammatory lung injury, and interstitial lung disease can produce overlapping findings. 8

Venous ultrasound examines a different consequence of filling. High right-sided pressure can be transmitted into hepatic, portal, and renal venous flow. The VExUS development study linked combined venous abnormalities to subsequent kidney injury after cardiac surgery. The findings can support concern about impaired drainage, but they do not calculate how much fluid a particular patient can safely receive. 9

Clinical trial evidence also shows why pulmonary consequences deserve attention. In FACTT, a conservative fluid strategy for acute lung injury increased ventilator-free days compared with a liberal strategy. The primary mortality outcome was not significantly different. Management during shock was handled separately, so the trial does not support withholding necessary resuscitation during persistent circulatory failure. 10

Look for agreement between findings. A rising filling pressure accompanied by worsening oxygenation supports a different explanation from isolated B-lines in a stable patient with chronic interstitial lung disease. Similarly, an apparently dry lung does not exclude renal or hepatic venous congestion.

Finally, consider timing and excretion. Pressure can rise before edema becomes visible. A patient with little initial congestion can develop it after retaining several further infusions. A reassuring assessment must therefore be repeated as treatment continues.

20.9 Combine Responsiveness with Tolerance

Responsiveness estimates whether more preload will increase flow. Tolerance estimates whether the accompanying pressure rise and fluid retention are likely to injure an organ. The two assessments can agree, or they can reveal a conflict that requires closer judgment.

This is a teaching framework, not a validated score. The quality of both assessments matters. An uninterpretable test should remain uncertain rather than being labeled negative.

The second row explains a common bedside dilemma. A patient may increase stroke volume during passive leg raising while already having severe lung injury or high ventricular filling pressures. The test establishes a possible flow benefit. It does not show that the lungs or venous circulation can accept the sustained infusion.

Choose alternatives according to the cause of inadequate circulation. Vasoplegia may require increased vascular tone. Contractile failure may require a different approach to cardiac support. Obstruction or excessive right ventricular afterload requires treatment of the limiting process. Each alternative also needs its own assessment of benefit and harm.

Nonresponsiveness does not automatically mean that fluid should be removed. It shows that adding preload is unlikely to increase flow under the tested conditions. Fluid removal requires a separate reason, such as clinically important accumulation or congestion, and adequate circulation while removal proceeds.

The aim is to improve the patient's perfusion problem while limiting organ injury. Neither maximizing stroke volume nor giving fluid until the patient becomes nonresponsive achieves that aim reliably.

20.10 What Clinical Trials Have Shown

A prediction study tests whether a maneuver forecasts an increase in flow. A treatment trial tests whether using that information improves a clinical outcome. A useful physiological test and a successful treatment strategy are related, but they require different evidence.

In FRESH, Douglas and colleagues used passive-leg-raise-induced stroke-volume changes to guide fluid and vasopressor treatment in sepsis-associated hypotension and shock. The strategy reduced net fluid balance. Some secondary organ-support outcomes also favored the intervention. The trial was modest in size and open label. Its results apply to the whole treatment strategy and cannot establish that the test or monitor alone produced the benefit. 11

CLASSIC compared restricted with standard intravenous fluid therapy in ICU patients with septic shock after initial fluid had already been given. Ninety-day mortality was not significantly different. The trial tested further fluid management at that stage of illness; it did not test whether patients should receive no initial resuscitation. 12

CLOVERS compared two early strategies after initial treatment for sepsis-induced hypotension. One emphasized less fluid and earlier vasopressors. The other gave more fluid before escalating vasopressors. Mortality before discharge home by day 90 was not significantly different. The comparison involved the timing and combination of treatments, rather than an isolated test of fluid against a vasopressor. 13

Melo and colleagues examined tolerance-based limits in a 2026 trial of fluid-responsive, ventilated patients with early septic shock and predefined vulnerability to fluid-related harm. Pulmonary, venous, and filling-pressure signals constrained further boluses. Recruitment stopped early: 36 of 62 planned patients were randomized, and 35 were analyzed. Protocol bolus volumes were lower, and the six-hour oxygenation change favored the intervention. Cumulative exposure including earlier fluid was similar. The small, open-label trial leaves the treatment effect uncertain and does not establish a mortality benefit or universal stopping thresholds. 14

The 2026 Surviving Sepsis Campaign guidelines conditionally suggest dynamic measures to guide fluid resuscitation over examination or static measures alone. The evidence is of low certainty. The recommendation supports using these tests within repeated clinical assessment; it does not turn every positive result into an instruction to give fluid. 15

A nonsignificant difference does not prove that the strategies are identical. The patient population, prior treatment, separation between groups, rescue treatments, and outcome measured determine what the trial can establish.

Keep the outcome explicit when discussing evidence. Predicting a stroke-volume increase, reducing fluid balance, shortening organ support, improving recovery, and reducing mortality are different claims. A study that establishes one does not automatically establish the others.

20.11 Apply the Principles at the Bedside

Consider an illustrative patient with septic shock and persistent peripheral hypoperfusion. A technically valid passive leg raise increases stroke volume beyond the threshold validated for the chosen method. Cardiac, pulmonary, and venous assessment identifies no major immediate concern about further fluid.

The result supports a preload-related component of the circulatory problem. If the history and recent treatment also support fluid, a defined challenge may be reasonable. Measure whether it increases flow, whether peripheral perfusion improves, and whether filling pressures or respiratory findings worsen.

Now consider a second illustrative patient with a similar positive flow response. This patient has severe pulmonary injury, rising cardiac filling pressures, and marked systemic venous abnormalities. The response is entirely possible: the heart can eject more blood while the additional filling worsens congestion.

For the second patient, identify what is limiting perfusion before choosing the next intervention. Review ventricular performance, vascular tone, pulmonary vascular load, ventilation, and surrounding pressures. If fluid remains necessary, define the exposure and stopping limits around the organs most likely to deteriorate.

Follow the result in both patients. An increase in stroke volume without improved perfusion may provide little benefit for the problem being treated. If perfusion improves briefly but oxygenation or congestion subsequently worsens, the initial positive response does not justify ignoring that harm.

Repeat the assessment after important changes. Additional fluid, a vasoactive adjustment, an arrhythmia, or a major ventilator change can alter preload responsiveness and tolerance. The earlier result cannot describe a circulation that has since changed.

At each decision, keep three judgments explicit. Establish why perfusion needs to improve. Determine whether additional preload is likely to improve it. Assess whether that benefit justifies the expected changes in filling pressures and tissue fluid. The next chapter examines vascular tone and vasoplegia, which often explain why fluid alone cannot correct the circulatory problem.

References

1. Roger C, Zieleskiewicz L, Demattei C, et al. Time course of fluid responsiveness in sepsis: the fluid challenge revisiting (FCREV) study. Crit Care. 2019;23:179. doi:10.1186/s13054-019-2448-z

2. Si X, Cao D, Shi R, et al. Fluid responsiveness and changes in venous congestion and lung water during volume expansion in critically ill patients: a multicentre observational study. Crit Care. 2026;30:35. doi:10.1186/s13054-025-05803-y

3. 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

4. Monnet X, Osman D, Ridel C, Lamia B, Richard C, Teboul JL. Predicting volume responsiveness by using the end-expiratory occlusion in mechanically ventilated intensive care unit patients. Crit Care Med. 2009;37(3):951–956. doi:10.1097/CCM.0b013e3181968fe1

5. Michard F, Boussat S, Chemla D, et al. Relation between respiratory changes in arterial pulse pressure and fluid responsiveness in septic patients with acute circulatory failure. Am J Respir Crit Care Med. 2000;162(1):134–138. doi:10.1164/ajrccm.162.1.9903035

6. Mahjoub Y, Pila C, Friggeri A, et al. Assessing fluid responsiveness in critically ill patients: false-positive pulse pressure variation is detected by Doppler echocardiographic evaluation of the right ventricle. Crit Care Med. 2009;37(9):2570–2575. doi:10.1097/CCM.0b013e3181a380a3

7. Cecconi M, Hofer C, Teboul JL, et al. Fluid challenges in intensive care: the FENICE study: a global inception cohort study. Intensive Care Med. 2015;41:1529–1537. doi:10.1007/s00134-015-3850-x

8. Lichtenstein D, Mézière G, Biderman P, Gepner A, Barré O. The comet-tail artifact: an ultrasound sign of alveolar-interstitial syndrome. Am J Respir Crit Care Med. 1997;156(5):1640–1646. doi:10.1164/ajrccm.156.5.96-07096

9. Beaubien-Souligny W, Rola P, Haycock K, et al. Quantifying systemic congestion with Point-Of-Care ultrasound: development of the venous excess ultrasound grading system. Ultrasound J. 2020;12:16. doi:10.1186/s13089-020-00163-w

10. National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome Clinical Trials Network; Wiedemann HP, Wheeler AP, Bernard GR, et al. Comparison of two fluid-management strategies in acute lung injury. N Engl J Med. 2006;354(24):2564–2575. doi:10.1056/NEJMoa062200

11. Douglas IS, Alapat PM, Corl KA, et al. Fluid response evaluation in sepsis hypotension and shock: a randomized clinical trial. Chest. 2020;158(4):1431–1445. doi:10.1016/j.chest.2020.04.025

12. Meyhoff TS, Hjortrup PB, Wetterslev J, et al. Restriction of intravenous fluid in ICU patients with septic shock. N Engl J Med. 2022;386(26):2459–2470. doi:10.1056/NEJMoa2202707

13. Shapiro NI, Douglas IS, Brower RG, et al. Early restrictive or liberal fluid management for sepsis-induced hypotension. N Engl J Med. 2023;388(6):499–510. doi:10.1056/NEJMoa2212663

14. Melo P, Amthauer M, Ferri G, et al. Fluid-intolerance signals as bedside safety limits during early septic shock resuscitation: a multicenter randomized trial. Crit Care. 2026;30:574. doi:10.1186/s13054-026-06316-y

15. Society of Critical Care Medicine and European Society of Intensive Care Medicine. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2026. Official recommendations