ICU · Renal / fluids
Fluid Overload & Its Management — The ROSE Concept & De-resuscitation
Also known as Fluid overload · Positive fluid balance · Cumulative fluid balance · Per cent fluid overload · De-resuscitation · The ROSE concept · SOSD · Fluid responsiveness · Fluid stewardship · The 4 Ds of fluid · Furosemide · CLASSIC trial · FACTT trial · FEAST trial · SMART trial · Global Increased Permeability Syndrome · Passive leg raise
The fluid overload is common in the ICU and is associated with the worse outcomes (the mortality, the prolonged ventilation, the AKI). It is quantified as the cumulative fluid balance (the net fluid in minus out) and the per cent fluid overload (the cumulative balance divided by the baseline weight times 100; the over 10 per cent is the worse). The harm: the pulmonary oedema, the tissue oedema (the gut, the kidneys, the intra-abdominal pressure), the delayed wound healing, the prolonged ventilation, the mortality. The ROSE concept frames the fluid therapy into the four phases — the Resuscitation, the Optimisation, the Stabilisation, and the De-resuscitation (the late phase that mobilises the excess fluid). The management: stop the fluids (minimise the maintenance), the de-resuscitation (the furosemide bolus or the infusion per the CLASSIC trial, the 20 per cent albumin, the RRT for the diuretic-resistant), the fluid-responsiveness assessment (the SVV, the PPV, the passive leg raise — NOT the CVP), and the fluid stewardship (the 4 Ds: the drug, the dose, the duration, the de-escalation).
On this page & tools
Your progress
Saved locally on this device.
8 MCQs with explanations
Target exams
Red flags
Overview & definition
The fluid overload is the common and the harmful complication of the ICU resuscitation. The cumulative fluid balance (the net fluid in minus out) rises during the resuscitation and, if not mobilised, causes the harm: the pulmonary oedema, the tissue oedema (the gut, the kidneys — the intra-abdominal pressure, the AKI), the delayed wound healing, the prolonged ventilation, and the mortality. The fluid overload is independently associated with the worse outcomes — it is a treatable, iatrogenic problem.[1]
Quantify the fluid overload
- The cumulative fluid balance — the net fluid in minus out (litres), summed over the ICU stay.[1]
- The per cent fluid overload = (the cumulative fluid balance / the baseline body weight) × 100. A per cent fluid overload over 10 per cent is associated with the worse outcomes (the mortality, the AKI, the prolonged ventilation).[1]
- The clinical signs: the peripheral oedema, the pulmonary oedema, the raised intra-abdominal pressure, the weight gain.[1]
The fluid balance is the algebraic sum of ALL the inputs (the IV fluids, the enteral, the parenteral nutrition, the drug diluents, the blood products) minus ALL the outputs (the urine, the drains, the NG aspirate, the fistula, the faeces, the insensible losses) from the ICU admission to the present. The daily weight (the bed scale) is the most objective measure — 1 kg of the weight change is about 1 L of the fluid, and the weight is superior to the balance chart (which misses the insensible losses, the fever, the open wounds, the drains). Target the return to the admission dry weight during the de-resuscitation.[1][4]
Why the fluid overload kills: the harms
The fluid overload harms the patient through the organ oedema, the raised tissue pressure, the impaired microvascular perfusion and the impaired oxygen diffusion. It is the iatrogenic, dose-dependent problem — the more the cumulative positive balance, the worse the outcome (the SOAP study showed the positive fluid balance is the independent predictor of the ICU mortality in the sepsis).[3]
Pulmonary harm
The most visible organ
- The interstitial and the alveolar oedema — the reduced compliance, the falling V/Q, the hypoxaemia, the rising plateau pressure for the same tidal volume
- The worsened oxygenation — the climbing FiO2 and the PEEP, the falling P/F ratio, the prolonged mechanical ventilation
- The FACTT (2006) — a conservative fluid strategy in the ALI/ARDS gave more ventilator-free days (14.6 vs 12.1) and more ICU-free days with no extra organ failure
- The lung ultrasound B-lines and the pleural effusions are the bedside signatures of the pulmonary congestion
Renal harm
A vicious cycle
- The renal venous congestion (the raised CVP transmitted back), the renal capsule stretch and the raised intra-abdominal pressure reduce the glomerular filtration rate
- The AKI then impairs the fluid excretion — more overload, more congestion, more AKI — a self-perpetuating spiral
- The fluid overload at the time of the RRT initiation is strongly associated with the worse renal recovery and the higher mortality
The gut and the abdomen
Often overlooked
- The bowel-wall oedema — the ileus, the impaired motility, the bacterial translocation, the intolerance of the enteral nutrition
- The third-spacing and the ascites raise the intra-abdominal pressure; an IAP over 20 mmHg with the new organ failure is the abdominal compartment syndrome (the urgent decompression)
- Measure the bladder pressure in any patient with a large fluid gain, a distended abdomen, the oliguria and the high peak airway pressures
The tissue, the wound and the brain
The diffuse burden
- The tissue oedema raises the wound tension and lowers the oxygen delivery — the impaired healing, the dehiscence, the surgical-site and the line infection, the pressure injury
- The cerebral oedema — even the modest fluid gains worsen the outcome in the traumatic brain injury and the hepatic encephalopathy
- The anasarca impedes the mobilisation and the rehabilitation and prolongs the ICU length of stay
The mortality signal
Independent and dose-dependent
- The SOAP study (Vincent 2006) — a positive cumulative fluid balance is the independent predictor of the ICU mortality in the sepsis
- The relationship is dose-dependent — each additional litre of the cumulative positive balance raises the mortality risk in the observational cohorts
- A per cent fluid overload above 10 per cent is the pragmatic threshold for the significant, harmful overload
The ROSE concept (the phases of the fluid therapy)
The fluid therapy has the four phases (the ROSE — or the SOSD: the Salvage, the Optimisation, the Stabilisation, the De-escalation):[1]
- The Resuscitation (the early, the salvage) — the fluid bolus for the life-threatening shock. Give enough to restore the perfusion.[1]
- The Optimisation — the titrated fluid for the cardiac output / the flow, guided by the fluid responsiveness.[1]
- The Stabilisation — the maintenance fluid, the minimal, the euvolaemia. Avoid the unnecessary fluid.[1]
- The De-resuscitation (the late, the goal-directed) — the mobilisation of the excess fluid once the shock has resolved. The diuretics or the RRT. The "late-and-goal-directed" approach.[1]
The fluid therapy in the critically ill is NOT a single act but a dynamic, time-dependent process. The cardinal error is failing to transition between the phases — continuing to give the fluid when the patient is no longer in the shock, or failing to de-resuscitate the accumulated fluid during the recovery. The target fluid balance CHANGES with the phase: POSITIVE (the resuscitation) → a diminishing positive (the optimisation) → ZERO (the stabilisation) → NEGATIVE (the de-resuscitation).[1][5]
The ROSE model: the four phases of the fluid therapy — with the targets and the timing
Phase 1 — Resuscitation (Salvage)
Timing: the minutes to the first few hours. Goal: reverse the life-threatening shock and restore the perfusion. Action: the rapid boluses of the balanced crystalloid (250–500 mL over 15–30 min; up to 30 mL/kg in the septic shock per the Surviving Sepsis Campaign). Give to any patient with the overt shock — the hypotension, the mottled skin, the altered mentation, the oliguria. The endpoints are macroscopic: a MAP above 65 mmHg, the falling lactate, the improving skin perfusion. The balance is intentionally POSITIVE. Do not delay the fluid for a dynamic test when the shock is obvious.
Phase 2 — Optimisation (Titration)
Timing: the hours 2 to 24. Goal: optimise the tissue perfusion while avoiding the overload. Action: give the fluid ONLY to the fluid-responsive patients, guided by the DYNAMIC indices — the passive leg raise, the PPV over 13 per cent, the SVV, the IVC collapsibility, the end-expiratory occlusion. The smaller titrated boluses (100–250 mL). Introduce the vasopressors early to reduce the total volume; add the inotropes for the low cardiac output. The endpoints shift to the micro level: an ScvO2 above 70 per cent, the lactate clearance above 10 per cent per hour, the urine output above 0.5 mL/kg/h. The balance is positive but with the diminishing gains.
Phase 3 — Stabilisation (Equilibrium)
Timing: from the day 2, once the shock has resolved. Goal: a ZERO or a slightly negative balance — neither gain nor loss. Action: STOP the maintenance fluids (the enteral nutrition provides most of the water). Eliminate the fluid creep — the concentrated drugs, the minimal flushes, no keep-vein-open. Track the daily weight, the IVC and the lung ultrasound. The patient should no longer need the boluses; the vasopressors are weaning. This is a transitional phase — do not linger here.
Phase 4 — De-resuscitation (Evacuation)
Timing: the recovery phase, the days to the weeks. Goal: a NEGATIVE balance — actively remove the accumulated fluid to return to the admission (dry) weight. Action: the furosemide IV (the bolus 20–40 mg titrated, or the continuous infusion 5–20 mg/h — more predictable and easier to titrate), add a thiazide (the metolazone 5–10 mg) for the sequential nephron blockade if the loop-resistant, the albumin 20 per cent for the oncotic pull if the hypoalbuminaemic, the CRRT for the diuretic-resistant or the oliguric. Target a net negative balance of 1–3 L/day, guided by the haemodynamic tolerance, the electrolytes and the daily weight.
The fluid responsiveness — before giving the fluid
Before giving the fluid (in the optimisation phase), assess the fluid responsiveness — the increase in the stroke volume with the fluid. Use the dynamic indices, NOT the static:[1]
- The stroke volume variation (SVV) and the pulse pressure variation (PPV) — the cardiopulmonary-interaction indices in the ventilated patient with the regular rhythm.[1]
- The passive leg raise (PLR) — the reversible endogenous fluid challenge; the increase in the stroke volume (or the cardiac output) of over 10 per cent indicates the responsiveness. Works in the spontaneous and the arrhythmic patient.[1]
- The end-expiratory occlusion test, the fluid challenge.[1]
- The static indices (the CVP, the IVC diameter alone) are poor predictors of the responsiveness — do not use them alone.[1]
The principle: only give the fluid if the heart is operating on the steep portion of the Frank-Starling curve — i.e., the ventricles are the preload-responsive. A fluid challenge to the non-responsive patient does NOT increase the cardiac output — it only increases the venous congestion, the oedema, and the harm. The static markers (the CVP, the heart rate, the blood pressure, the capillary refill, the lactate, the urine output) are the poor predictors of the responsiveness alone — the dynamic tests (the PLR, the PPV/SVV, the IVC, the EEO) are always superior.[1][7]
Passive leg raise (PLR)
The best all-round test
- A reversible endogenous fluid challenge — elevating the legs to 45 degrees auto-transfuses about 300 mL of the venous blood from the legs and the splanchnic bed
- Positive if the stroke volume or the cardiac output rises by more than 10 per cent (measure with the echocardiographic LVOT VTI, the pulse-contour analysis, or the pulse pressure from the arterial line)
- Works in the spontaneously breathing and the arrhythmic patient — its great advantage over the PPV/SVV
- Start from a semi-recumbent position; lower the trunk and raise the legs. Avoid in the leg amputation or the compression; it needs a real-time cardiac output measurement, not the blood pressure alone
PPV and SVV
Best for the controlled ventilated patient
- In the controlled mechanical ventilation, the inspiration reduces the right heart preload and raises the afterload, producing the cyclic variation in the left ventricular stroke volume and hence in the pulse pressure
- A PPV above 13 per cent (or an SVV above 10 per cent) predicts the fluid responsiveness with the good sensitivity and the specificity
- Requirements: the fully controlled ventilation (no spontaneous breaths), the tidal volume above 8 mL/kg PBW, the sinus rhythm, the closed chest, no severe right heart failure
- Invalid in: the spontaneous breathing, the atrial fibrillation, the low-tidal-volume ventilation, the open chest, the severe RV failure, the very low lung compliance
IVC collapsibility / distensibility
The simplest bedside test
- The subcostal POCUS — measure the IVC diameter and its change with the respiration
- In the spontaneously breathing patient: a collapsibility index above 50 per cent (a small, snapping IVC) suggests the volume responsiveness; a plethoric fixed IVC (the collapsibility below 50 per cent, the diameter above 2.5 cm) suggests the overload — do NOT give the fluid
- In the ventilated patient use the distensibility — an inspiratory increase above 18 per cent suggests the responsiveness
- Quick, non-invasive, repeatable — but semi-quantitative and operator-dependent
End-expiratory occlusion (EEO)
For the ventilated patient with the arrhythmia
- Hold the expiration for 15 seconds — this removes the intrathoracic pressure swing, transiently increasing the venous return; a rise in the arterial pulse pressure (or the cardiac output) above 5 per cent predicts the responsiveness
- Validated by the Monnet (2009) — useful where the PPV/SVV fail (the arrhythmia, the spontaneous breathing effort on the ventilator)
- Needs an arterial line and a cooperative ventilator; the change is small so the direct cardiac output measurement (the echocardiography) improves the detection
Mini-fluid challenge
A real, small bolus
- 50–100 mL of the colloid given rapidly over one minute; a rise in the cardiac output (the LVOT VTI) of 10–15 per cent indicates the responsiveness
- Smaller and safer than a 250–500 mL challenge — it minimises the harm if the patient is not responsive
- Use a directly measured stroke volume, not the blood pressure; less useful without the echocardiography or the pulse-contour output
Static indices (limited)
Do NOT use alone
- The CVP (the central venous pressure) is a POOR predictor of the fluid responsiveness — the classic Marik data shows almost no relationship between the CVP and the response to a bolus
- The heart rate, the blood pressure and the capillary refill are the late and the non-specific markers — the tachycardia may be the pain, the fever or the anxiety
- The lactate marks the hypoperfusion, not the volume status — an elevated lactate needs the investigation, not an automatic bolus
- Principle: the dynamic tests (the PLR, the PPV, the IVC, the EEO) are always superior to the static markers for the fluid responsiveness
The management
1. Stop the fluids
- The first step. Minimise the maintenance fluids, the drug diluents, the keep-vein-open. Use the concentrated medications, the enteral route.[1]
- The balanced crystalloids (not the chloride-liberal) — the balanced solutions reduce the AKI.[1]
- The early enteral nutrition to reduce the IV fluid calories.[1]
2. De-resuscitate (the mobilisation of the excess)
- The furosemide — the loop diuretic, the bolus or the infusion. The CLASSIC trial showed the bolus furosemide strategy in the fluid-overloaded ICU patient was safe and achieved the negative balance. The dose titrated to the fluid-balance target. The bolus-vs-infusion and the albumin adjunct.[1]
- The albumin 20 per cent — the colloid, the oncotic pull of the fluid from the interstitium into the vasculature (the adjunct to the diuretic).[1]
- The RRT (the slow continuous ultrafiltration, the CVVH) — for the oliguric or the diuretic-resistant fluid overload. The isolated ultrafiltration for the pure fluid removal.[1]
3. The fluid stewardship (the 4 Ds)
The fluid treated as a drug — the 4 Ds: the drug (the balanced crystalloid), the dose (the right amount), the duration (the shortest), and the de-escalation (the daily review, the minimise). The daily review of the fluid balance and the weight.[1]
De-resuscitation: the stepwise toolkit for the fluid removal
Step 1 — Stop the fluid input
The simplest and the most effective intervention. Audit EVERY IV line: stop the maintenance fluids (the enteral nutrition provides the water), concentrate all the drug infusions (the noradrenaline in 50 mL not 250 mL, the double-strength preparations), minimise the flush volumes, and challenge every blood-product order (do not transfuse for the numbers alone). A pharmacist-led fluid-stewardship ward round can remove 1–2 L/day of the unnecessary input.
Step 2 — Furosemide (the loop diuretic), the first line
Start with 20–40 mg IV bolus; assess the response in 30–60 minutes (the urine output should rise). If inadequate, double the dose (40 then 80 then 120 mg). For the sustained removal a continuous infusion (5–20 mg/h) is more predictable, less ototoxic and easier to titrate than the repeated boluses. Target a net negative balance of 1–3 L/day. Monitor the electrolytes (the potassium, the magnesium, the calcium), the creatinine (avoid the over-diuresis into the prerenal AKI), the blood pressure and the urine output.
Step 3 — The sequential nephron blockade (add a thiazide)
For the loop-diuretic resistance (common in the chronic kidney disease, the heart failure and the hypoalbuminaemia), add a thiazide to block the distal convoluted tubule: the metolazone 5–10 mg via the NG (preferred — potent even in the CKD), the chlorthalidone, or the IV chlorothiazide. The combination is synergistic and can produce the MASSIVE diuresis — start low (the metolazone 2.5–5 mg) and check the potassium and the magnesium every 6–12 hours; replace aggressively.
Step 4 — Albumin 20 per cent (the oncotic therapy)
In the hypoalbuminaemic patient (the albumin below 25 g/L), the fluid leaks into the interstitium and the furosemide is less effective (it must bind the albumin to reach the tubular site of the action). The albumin 20 per cent (100–200 mL) followed by the furosemide raises the oncotic pressure, pulls the fluid back into the vascular space and potentiates the diuresis. The SAFE trial showed the albumin and the saline are equivalent for the resuscitation — use the albumin selectively as the oncotic adjunct, not for the routine volume expansion.
Step 5 — The CRRT for the refractory overload
When the diuretics fail or the concurrent AKI prevents a diuresis, start the CRRT (the CVVHDF or the CVVH) for the slow, controlled, haemodynamically stable fluid removal. Set a net ultrafiltration target (start at −50 to −100 mL/h, about −1.2 to −2.4 L/day) and increase as the blood pressure tolerates. The CRRT also clears the urea, the creatinine and the potassium. The CRRT for the fluid overload alone (without the uraemia) is justified when the diuretics fail.
Step 6 — Monitor and titrate
Track the response DAILY: the weight (aim a downward trend toward the dry weight), the fluid balance (confirm the net negative), the IVC (should become more collapsible), the lung ultrasound (the B-lines should clear) and the oxygenation. Watch for the over-diuresis — the hypotension, the rising lactate, the worsening AKI, the electrolyte derangement. The endpoint is the return to the admission weight, the resolution of the overload signs and the improved organ function.
Drug — the right fluid
The type matters
- The balanced crystalloids (the Plasma-Lyte, the Hartmann, the Ringer lactate) preferred over the 0.9 per cent saline — the SMART trial showed the fewer major adverse kidney events and the lower mortality with the balanced solutions
- Avoid the chloride-liberal fluids — the hyperchloraemic metabolic acidosis, the renal vasoconstriction and the AKI
- The colloids (the albumin 4 per cent) are equivalent to the saline for the resuscitation (the SAFE); the albumin 20 per cent is the oncotic adjunct, not the routine resuscitant
- The starches are harmful — the increased AKI and the mortality; do not use
Dose — the right amount
The bolus only for the shock
- A bolus is 250–500 mL of the balanced crystalloid over 15–30 minutes, given only for the overt shock or a positive dynamic test
- Up to 30 mL/kg in the first hours of the septic shock (the Surviving Sepsis Campaign) — but titrated, never as a single blind bolus
- The FEAST — the routine bolus therapy in the patients without the hypovolaemic shock was harmful; the fluid is a drug with a narrow therapeutic window
Duration — the shortest course
Stop early
- Stop the maintenance fluids as soon as the enteral nutrition is tolerated — the enteral feeding provides the daily water requirement
- Concentrate the drug infusions; review the flush volumes and the carrier fluids daily
- The CLASSIC — a restrictive strategy in the septic shock (the mean about 1.8 L vs about 3.8 L over the ICU stay) was safe; less fluid, less overload
De-escalation — the daily review
Prevent the accumulation
- The daily review of the cumulative fluid balance, the per cent fluid overload and the weight — at the bedside, every round
- Move from the positive to the zero to the negative balance as the patient recovers (the ROSE transition)
- Plan and execute the de-resuscitation once the shock resolves — do not let a positive balance accumulate
The key trials in the fluid therapy
CLASSIC — Conservative vs Liberal fluid therapy of Septic Shock (Meyhoff, NEJM 2022)
International, multicentre, open-label RCT; 1554 adults with the septic shock across nine ICUs
Population: The ICU patients with the septic shock who had already received at least 1 L of the IV fluid
Key finding
No significant difference in the 90-day mortality (42.3 per cent restrictive vs 42.1 per cent standard; adjusted RR 1.00, 95 per cent CI 0.89–1.13, p=0.96). The restrictive group received far less IV fluid (the mean 1798 mL vs 3811 mL during the ICU stay) with no increase in the serious adverse events. A pre-specified subgroup analysis signalled the benefit with the restriction in the less severe shock and in the patients without the metastatic cancer.
Practice change
A restrictive fluid strategy in the septic shock is SAFE — it did not raise the mortality despite giving about 2 L less fluid. The trial was underpowered for a modest effect, but the signal favours the restriction. This is the definitive modern evidence supporting the fluid stewardship: less fluid means less overload and less harm, with the equivalent outcomes.
FACTT — Fluid and Catheter Treatment Trial (Wiedemann, NEJM 2006)
Multicentre RCT; 1000 patients with the acute lung injury (the ARDSNet)
Population: The mechanically ventilated adults with the acute lung injury
Key finding
No significant difference in the 60-day mortality (25.5 per cent vs 28.4 per cent, p=0.30). The conservative strategy gave MORE ventilator-free days (14.6 vs 12.1, p<0.001), MORE ICU-free days (13.4 vs 11.2, p=0.003), the improved oxygenation, with NO increase in the non-pulmonary organ failure and NO increase in the shock.
Practice change
A conservative fluid strategy in the ALI/ARDS is safe and beneficial — it shortens the ventilation and the ICU stay without causing the harm. The fluid overload worsens the pulmonary outcomes; once the shock resolves, target a negative or an even balance.
FEAST — Fluid Expansion As Supportive Therapy (Maitland, NEJM 2011)
Multicentre open-label RCT; 3170 children in the Uganda, the Kenya and the Tanzania
Population: The children (60 days to 12 years) with the severe febrile illness and the impaired perfusion in a resource-limited setting
Key finding
The BOLUS INCREASED MORTALITY. The 48-hour mortality was 10.5 per cent (the saline bolus) and 10.6 per cent (the albumin bolus) vs 7.3 per cent (no bolus). The relative risk of the death with a bolus was 1.45 (95 per cent CI 1.13–1.86, p=0.003). The trial was STOPPED EARLY for the harm — the mechanism was the cardiogenic pulmonary oedema, the intracranial hypertension and the haemodynamic compromise in the children without the true hypovolaemic shock.
Practice change
A fluid bolus is NOT benign — in the patients without the true hypovolaemic shock it can be HARMFUL. This landmark trial reshaped the paediatric and the adult resuscitation thinking: assess the fluid responsiveness before any bolus and never give the fluid reflexively. The fluid is a drug with a therapeutic window — too little is dangerous, but too much is equally dangerous.
SMART — Balanced Crystalloids vs Saline in Critically Ill Adults (Self, NEJM 2018)
Single-centre, cluster-crossover, pragmatic RCT; 15,752 adults across five ICUs
Population: All the adults admitted to a medical, surgical, cardiac, trauma or neurosciences ICU
Key finding
The balanced crystalloids reduced the MAKE30 (14.3 per cent vs 15.4 per cent; OR 0.90, 95 per cent CI 0.82–0.99; p=0.04). The benefit was larger in the sepsis subgroup (the in-hospital mortality 25.2 per cent vs 28.7 per cent).
Practice change
The balanced crystalloids are preferred over the saline for the routine ICU fluid therapy — the fewer major adverse kidney events. The Drug of the 4 Ds of the fluid stewardship should be a balanced crystalloid, not a chloride-liberal saline.
SAFE — Saline vs Albumin Fluid Evaluation (Finfer, NEJM 2004)
Multicentre, double-blind RCT; 6997 critically ill adults
Population: All the ICU patients needing the fluid resuscitation
Key finding
No difference in the 28-day mortality (20.9 per cent albumin vs 21.1 per cent saline; RR 0.99, 95 per cent CI 0.91–1.09, p=0.87). The outcomes were equivalent across the subgroups.
Practice change
For the routine fluid resuscitation, the 4 per cent albumin and the saline are equivalent — the albumin is not superior and is more expensive. The albumin 20 per cent (a different product) retains the role as the oncotic adjunct in the de-resuscitation of the hypoalbuminaemic patient, not as the routine resuscitant.
SAQs
SAQ — Fluid overload assessment in septic shock
10 minutes · 10 marks
A 68-year-old man (70 kg) is on day 4 of ICU admission for septic shock from a perforated viscus. He has received a cumulative balance of +9 L of balanced crystalloid. He is ventilated (FiO2 0.5, PEEP 10) with rising plateau pressure, oliguric (0.2 mL/kg/h), on noradrenaline 0.15 mcg/kg/min, and his lactate has cleared to 1.4 mmol/L. The team asks you to assess his fluid status and fluid responsiveness.
SAQ — De-resuscitation of the recovering critically ill patient
10 minutes · 10 marks
A 55-year-old woman (80 kg) is on day 6 of ICU admission for severe acute pancreatitis complicated by septic shock. She is now off vasopressors for 24 hours, with a lactate of 1.1 mmol/L and a MAP of 75 mmHg without support. Her cumulative fluid balance is +12 L (15 per cent of body weight). She remains ventilated (FiO2 0.4, PEEP 8) with oxygenation deteriorating, is oliguric (0.3 mL/kg/h), has sacral and pedal oedema, and her serum creatinine has risen from 90 to 220 micromol/L. You are asked to de-resuscitate her.
Clinical pearls
Prognosis
[1]Red flags
References
- [1]Maitland K, Kiguli S, Opoka RO, et al. Mortality after fluid bolus in African children with severe infection N Engl J Med, 2011.PMID 21615299
- [2]Wiedemann HP, Wheeler AP, Bernard GR, et al. (ARDSNet/FACTT) Comparison of two fluid-management strategies in acute lung injury N Engl J Med, 2006.PMID 16714767
- [3]Vincent JL, Sakr Y, Sprung CL, et al. (SOAP Study) Sepsis in European intensive care units: results of the SOAP study Crit Care Med, 2006.PMID 16424713
- [4]Cordemans C, De Laet I, Van Regenmortel N, et al. Fluid management in critically ill patients: the role of extravascular lung water, abdominal hypertension, capillary leak, and fluid balance Ann Intensive Care, 2012.PMID 22873410
- [5]Silversides JA, Fitzgerald E, Manu VS, et al. Conservative fluid management or deresuscitation for patients with sepsis or acute respiratory distress syndrome following the resuscitation phase of critical illness: a systematic review and meta-analysis Intensive Care Med, 2017.PMID 27734109
- [6]Meyhoff TS, Hjortrup PB, Wetterslev J, et al. (CLASSIC) Restriction of Intravenous Fluid in ICU Patients with Septic Shock N Engl J Med, 2022.PMID 35709019
- [7]Monnet X, Osman D, Ridel C, et al. Predicting volume responsiveness by using the end-expiratory occlusion in mechanically ventilated intensive care unit patients Crit Care Med, 2009.PMID 19237902
- [8]Finfer S, Bellomo R, Boyce N, et al. (SAFE Study) A comparison of albumin and saline for fluid resuscitation in the intensive care unit N Engl J Med, 2004.PMID 15163774
- [9]Self WH, Semler MW, Wanderer JP, et al. (SMART) Balanced Crystalloids versus Saline in Critically Ill Adults N Engl J Med, 2018.PMID 29768150