A reversible "self-fluid-challenge" — auto-transfusing venous blood centrally to predict who will actually benefit from a fluid bolus, before a single mL is infused
Roughly half of hemodynamically unstable patients in the ICU or operating room do not increase their cardiac output in response to a fluid bolus — yet clinicians have historically given fluid empirically, guided by static pressures like CVP that predict almost nothing about the heart's actual position on the Frank-Starling curve. Passive leg raise (PLR) testing was developed to answer one question before committing a single milliliter: will this specific heart, at this specific moment, actually respond?
For decades, central venous pressure (CVP) and pulmonary artery occlusion pressure were used to decide whether a patient "needed" fluid. Both are static, single-point measurements of pressure — but the relevant physiological question is about position on the Frank-Starling curve, a dynamic relationship between preload and stroke volume that a single pressure number cannot reveal.
A patient with a high CVP may still be fluid-responsive (steep curve, but venoconstricted or on a ventilator with high intrathoracic pressure); a patient with a low CVP may already be flat on the curve. Pooled data across dozens of studies show CVP has essentially no discriminative value for predicting the hemodynamic response to fluid (AUC close to 0.55, i.e., near coin-flip).
This matters enormously in practice: every unnecessary liter of fluid pushed into a non-responder does nothing for perfusion and instead accumulates as edema — in the lungs, the gut wall, the kidneys, and the skin.
Under-resuscitation leaves a genuinely fluid-responsive patient hypoperfused: organs starved of oxygen delivery, rising lactate, worsening acute kidney injury, progression toward shock.
Over-resuscitation of a non-responder is equally dangerous but slower to declare itself: interstitial edema impairs oxygen diffusion at the very tissue level fluid was meant to help, raises intra-abdominal pressure, worsens lung compliance and ventilator-free days, and is independently associated with increased mortality in septic shock and ARDS cohorts.
The clinical need, therefore, is a test that is fast, reversible, bedside-deployable, and does not itself commit fluid before the answer is known — which is exactly what passive leg raise provides.
Other dynamic indices exist — pulse pressure variation (PPV), stroke volume variation (SVV) — which exploit heart-lung interactions during positive-pressure ventilation. These are accurate but require strict conditions: fully controlled mechanical ventilation, tidal volume ≥8 mL/kg, sinus rhythm, closed chest, no spontaneous breathing effort.
PLR sidesteps all of these prerequisites. It works in spontaneously breathing patients, in arrhythmia, in low tidal volume ventilation, and in patients with an open abdomen — precisely the populations where PPV/SVV are unreliable. It achieves this because it does not rely on ventilator-induced preload cycling at all; it physically moves blood using gravity and repositioning.
The central insight of PLR is elegant: instead of asking "what does this pressure number suggest," we perform an actual, reversible mini fluid challenge using the patient's own blood volume — then measure the direct hemodynamic answer in real time.
The PLR maneuver is deceptively simple: starting from a 45° semi-recumbent position, the bed is tilted so the trunk moves to horizontal and the legs are raised to 45° — all in a single, smooth motion, ideally using the bed mechanism itself rather than manually lifting the limbs. This repositioning drains venous blood from the legs and splanchnic reservoir into the thorax, transiently increasing cardiac preload by roughly 300 mL without opening an IV line.
The maneuver must start from the semi-recumbent position (trunk at 45°), not from fully supine. Starting supine and merely lifting the legs only mobilizes calf and thigh venous volume; starting semi-recumbent and moving to trunk-flat-plus-legs-raised also drains the large splanchnic venous reservoir, roughly doubling the effective preload challenge.
The preferred method uses the bed's own hinge mechanism: the head-of-bed motor lowers the trunk to horizontal at the same time the leg segment raises to 45°, in one continuous automated motion — this avoids the sympathetic/pain-mediated norepinephrine surge that can confound results if staff physically lift the patient's legs by hand.
Because the maneuver mobilizes a fixed, endogenous blood volume rather than external fluid, it is often described as testing the response to "auto-transfusion" — the patient literally gives themselves a temporary bolus from their own venous reservoirs.
Roughly 70% of total blood volume resides in the venous system at any moment, and the legs and splanchnic bed together hold a substantial fraction of that as a compliant, low-pressure reservoir. Gravitational repositioning shifts an estimated 150–300 mL of this reservoir toward the right atrium within seconds.
This preload increase is genuinely comparable in magnitude to a 250–500 mL crystalloid bolus given over 10–15 minutes — but delivered in under a minute, and, crucially, fully reversible the instant the position is reverted. No capillary leak, no third-spacing, no committed fluid balance.
Because venous capacitance and compliance vary between patients (vasoplegia in septic shock increases capacitance; vasoconstriction reduces it), the exact volume mobilized differs, but the directionality and interpretive logic of the test remain the same across populations.
Several technical errors reduce or invalidate the PLR signal:
• Manually lifting only the legs from full supine (skips the splanchnic reservoir contribution) • Applying pain or agitation during positioning, which triggers a sympathetic surge that masks or mimics the hemodynamic response • Failing to allow the patient to return to baseline between repeated tests (residual preload effect contaminates the next reading) • Measuring cardiac output with a slow or averaged method that cannot capture the transient peak within the 1–2 minute effective window
When performed correctly and measured with a fast, continuous monitor, PLR is one of the most reproducible bedside hemodynamic tests available.
Because the PLR effect on preload is transient — rising within seconds and beginning to wane after roughly 1–2 minutes as compensatory mechanisms and venous re-equilibration set in — the measurement tool must be fast and continuous. A slow, intermittent method (like a manual cuff blood pressure every few minutes) will simply miss the window entirely and produce a false-negative result.
The gold-standard bedside monitor for PLR testing is a continuous cardiac output device derived from arterial pulse-contour analysis (e.g., PiCCO, FloTrac/Vigileo, LiDCO, or a similar beat-to-beat algorithm). These devices calculate stroke volume from the shape and area of the arterial pressure waveform on every single heartbeat, updating a rolling cardiac output value several times per second.
This granularity is essential: it allows the clinician to watch stroke volume climb in real time as venous return increases during the leg-raise, identify the true peak (rather than a single delayed snapshot), and confirm the value returns toward baseline once the legs are lowered — which itself is a useful internal consistency check that the initial rise was a genuine hemodynamic response rather than measurement noise.
When an arterial line and pulse-contour device are unavailable, several validated surrogates can substitute:
• Transthoracic or transesophageal echocardiography: measuring the aortic or left ventricular outflow tract velocity-time integral (VTI) before and during PLR; a ≥10% rise in VTI mirrors the stroke volume threshold • End-tidal CO2 (EtCO2): in mechanically ventilated patients with stable ventilation, EtCO2 tracks cardiac output because pulmonary blood flow determines CO2 delivery to the alveoli for exhalation; a rise of ≥5% during PLR has reasonable predictive value • Bioreactance / non-invasive cardiac output monitors: thoracic bioimpedance-based devices offer a fully non-invasive continuous CO trend, useful outside the ICU
Whatever the modality, the requirement is the same: fast enough temporal resolution to capture a transient signal that rises and falls within roughly 90 seconds.
A common bedside shortcut is to watch arterial blood pressure instead of directly measuring cardiac output — but blood pressure is the product of cardiac output AND vascular resistance, and PLR can transiently alter both. A patient with poor arterial compliance may show a brisk pressure rise with little true flow increase, while a vasodilated septic patient may show a genuine flow increase with a disappointingly flat pressure trace.
Guidelines therefore recommend that whenever feasible, the response should be judged on a direct flow-based variable — stroke volume or cardiac output — rather than pressure alone, reserving pulse pressure only as a rough, lower-confidence fallback when no flow monitor exists.
The measurement device is the rate-limiting step of the whole test: a physiologically perfect maneuver paired with a monitor too slow to see the 60–90 second window will always produce a falsely reassuring "non-responder" result.
The entire clinical value of PLR collapses into a single decision threshold: does cardiac output or stroke volume rise by roughly 10% or more during the maneuver? Above that threshold, the patient sits on the ascending (steep) portion of the Frank-Starling curve — more preload genuinely buys more output. At or below it, the patient has already reached the flat portion of the curve, where additional preload produces negligible gain in stroke volume, only added volume the body must eventually clear or sequester as edema.
The Frank-Starling law describes how stroke volume rises with increasing preload (end-diastolic volume/stretch) up to a physiological limit, beyond which further preload produces diminishing and eventually flat returns. Two patients can have identical filling pressures yet sit at entirely different points on this curve depending on ventricular contractility, compliance, and afterload.
PLR effectively performs a small, controlled, reversible preload challenge and reads the corresponding stroke volume change — directly sampling the local slope of that patient's own curve at that specific moment, rather than inferring position from a static pressure number.
Because the curve's shape is patient-specific and changes over the course of an illness (a patient can move from responsive to non-responsive as resuscitation progresses), the test is meant to be repeated at decision points, not performed once and assumed to hold indefinitely.
A ≥10% rise in CO/SV during PLR indicates the ventricle is preload-dependent: giving an actual fluid bolus is expected to meaningfully increase cardiac output and, downstream, oxygen delivery. This is the scenario where committing real intravenous fluid is justified by objective evidence rather than a reflexive response to hypotension or oliguria.
Even here, PLR only answers "will output rise" — it does not by itself confirm that the patient needs more output (a patient can be preload-responsive yet already adequately perfused). It should be paired with clinical assessment of perfusion (lactate, capillary refill, urine output, mixed venous saturation) to decide whether responsiveness should be acted upon.
A rise below the ~10% threshold (including flat or negative changes) indicates the patient is already on the flat part of the curve: further fluid is very unlikely to improve cardiac output meaningfully, and the identical volume that would have been infused instead accumulates as tissue edema with no compensating hemodynamic benefit.
In this scenario, hemodynamic instability is better addressed through alternative strategies: vasopressor support to restore perfusion pressure without added volume, inotropic support if a primary contractility problem is suspected, or simply re-examining the diagnosis (is this bleeding, tamponade, obstructive shock, or a source needing source control rather than volume?).
Interpreted correctly, a negative PLR is just as clinically actionable as a positive one — it is the evidence needed to stop reflexively bolusing fluid and pivot the resuscitation strategy toward vasopressors, inotropes, or diagnostic reassessment.
The single feature that separates PLR from every real fluid bolus is that its effect disappears the moment the legs are lowered back down. The auto-transfused venous volume simply redistributes back to the periphery, and hemodynamics return to their pre-maneuver baseline within roughly one to two minutes — no capillary leak has occurred, no fluid has been committed to the interstitium, and nothing needs to be "cleared" afterward.
Traditional fluid-challenge strategies infuse a real bolus (commonly 250–500 mL of crystalloid) and observe the response — but if the patient turns out to be a non-responder, that volume is now a liability rather than a diagnostic tool: it has already left the intravascular space and begun contributing to interstitial edema, and it cannot be "un-given."
PLR inverts this risk entirely. Because the tested volume is the patient's own blood, temporarily repositioned rather than infused, a negative test costs nothing beyond the two minutes it took to perform. The clinician gains the diagnostic information of a fluid challenge without incurring the physiological debt of one.
Because PLR carries essentially no downside, it can be repeated as often as clinically useful — before and after a vasopressor titration, after a change in ventilator settings, after a real fluid bolus has been given (to confirm the patient has moved off the steep part of the curve), or simply at routine intervals during an evolving shock state.
This repeatability transforms fluid management from a series of one-way, irreversible decisions into an iterative feedback loop: test, decide, treat, retest. Over a resuscitation course, a patient will typically transition from responsive to non-responsive as effective circulating volume is restored — and PLR is one of the few bedside tools that can track that transition safely, in real time, without adding cumulative fluid balance risk of its own.
PLR is not universally applicable. Several conditions reduce the reliability or feasibility of the maneuver:
• Intra-abdominal hypertension: elevated intra-abdominal pressure can impede the venous return augmentation that PLR depends on, blunting the true response and risking a false negative • Raised intracranial pressure: leg elevation and trunk repositioning can transiently raise ICP; caution or an alternative test is preferred in severe traumatic brain injury • Lower-limb amputation or compression devices that prevent proper leg elevation • Significant arrhythmia during the measurement window, which can confound beat-to-beat pulse-contour readings independent of the preload effect
Recognizing these caveats is part of using PLR responsibly: like any bedside test, a technically flawed or contextually inappropriate application can produce a misleading result, and clinical judgment must always frame the number.
Reversibility is not a convenience feature — it is the property that makes PLR ethically and physiologically safe to use liberally, turning a single high-stakes fluid decision into a low-risk, repeatable diagnostic loop.