De-resuscitation in Intensive Care

The accumulation of a positive fluid balance is a frequent occurrence in critically ill patients. As Paracelsus said, ” All things are poisons, nothing is without poison. Only the dose permits something not to be poisonous.” Like other drugs, it is the dose of fluids which makes them poisonous.

What is de-resuscitation? Where did this term originate from?
The use of intravenous fluid therapy is universal among critically ill patients to optimize tissue perfusion and oxygen delivery. Apart from intravenous fluids administered during initial resuscitation, fluid accumulation occurs from nutrition, maintenance fluids, and diluents used for intravenous drug therapy. An aggressive resuscitation strategy may be appropriate during the “ebb” phase of septic shock, characterized by widespread vasodilatation, intravascular hypovolemia, and capillary leak, leading to extravasation of fluid into the interstitial compartment. Among patients who respond to initial therapeutic interventions, shock resolution and circulatory stabilization typically occur. By this time, a large positive fluid balance occurs among critically ill patients, particularly among those with sepsis. In a secondary analysis of the Vasopressin in Septic Shock Trial (VASST), the mean positive fluid balance was 4.2 L at 12 hours and 11.0 L on the 4^th day of enrollment. It is important to take stock of the net fluid balance after the initial phase of resuscitation. The extent of fluid accumulation may be calculated using the formula: 

[Cumulative fluid balance (liters) / baseline body weight (kg)] × 100

A fluid accumulation of more than 10% has been shown to be associated with poor clinical outcomes.  During the “flow” phase, the body attempts to dispose of the accumulated fluid. Therapeutic fluid removal, using diuretics and ultrafiltration with renal replacement therapy, may be appropriate measures during this stage. Aggressive interventions to remove fluid may be combined with a fluid-restrictive strategy. These interventions aimed at active fluid removal after the initial phase of resuscitation has been alluded to as “de-resuscitation”. The term “de-resuscitation” was coined by Malbrain et al. in their systematic review along with therapeutic recommendations in 2014.

What’s the rationale for this?

While an aggressive resuscitation strategy may be appropriate during the “ebb” phase of septic shock, characterized by widespread vasodilatation, intravascular hypovolemia, and capillary leak, leading to extravasation of fluid into the interstitial compartment. Typically, the “ebb” phase is followed by the “flow” phase, during which the body tries to eliminate excess fluid. However, normal physiological mechanisms may be overwhelmed, especially with the onset of acute kidney injury. Excessive fluid accumulation may trigger multiorgan dysfunction. Interstitial and intra-alveolar edema in the lung may lead to delayed weaning and prolongation of the duration of ventilator support. A rise in the venous pressure may impair renal blood flow, trigger interstitial edema, and result in compromised renal function. Myocardial edema may occur, leading to systolic and diastolic dysfunction. As the intra-abdominal pressure rises due to excessive gastrointestinal fluid accumulation, abdominal perfusion may be compromised. This may trigger increased intestinal permeability and bacterial translocation from the gut. The potential harm arising from injudicious fluid administration has been equated to adverse effects arising from inappropriate drug dosing. 

What’s the evidence either in favour of or against it?
Fluids administered during the early resuscitation phase may constitute only a small fraction of the fluid accumulated over time. A retrospective analysis was performed to evaluate the volume of fluids administered to 14,654 critically ill patients. In this study, maintenance and replacement fluids constituted 24.7% of the daily fluid administered, compared to only 6.5% of resuscitation fluids. Significant volumes of fluid (32.6% of the daily volume administered) were administered as diluents for intravenous or other medication. 

Silversides et al. conducted a retrospective observational study across 10 ICUs in the United Kingdom and Canada to identify outcomes with de-resuscitative strategies and risk factors for positive fluid balance among critically ill patients. Four-hundred adult patients who underwent mechanical ventilation for a minimum period of 24 hours were included. Over the first 1–3 days, a positive fluid balance occurred in 87.3% of patients. Medications represented the largest volume of fluid administered (34.5%). The volume of bolus (26.5%) and maintenance fluids (24.4%) were lower compared to fluids administered as medication. A negative balance was observed on day 3 in 123 patients; a spontaneous negative balance, without the use of diuretics or ultrafiltration, occurred in 70 (56.9%) patients and was associated with lower 30-day mortality. De-resuscitative measures were employed in 52.3% of all patients, most frequently on the 2^nd or 3^rd day. On multivariate logistic regression, the fluid balance on day 3 was an independent predictor of 30-day mortality. A negative fluid balance achieved using active de-resuscitative measures, including frusemide administration and fluid removal using continuous renal replacement therapy, was associated with lower mortality. 

A meta-analysis of 11 randomized controlled studies, including 2051 adults and children, was performed by Silversides et al. The study included patients with acute respiratory distress syndrome, sepsis, or the systemic inflammatory response syndrome (SIRS). Compared to a liberal approach, the authors found no significant difference in mortality reported at the longest time point between a conservative or de-resuscitative fluid management strategy. However, there was an increase in ventilator-free days and reduced length of stay in the ICU with a conservative or de-resuscitative approach. 

The FFAKI-trial was a pilot randomized controlled trial that tested the feasibility of forced fluid removal compared to standard care among critically ill patients who were at moderate to high risk of acute kidney injury with more than 10% fluid accumulation. Active de-resuscitation was carried out with a bolus dose of furosemide, 40 mg, followed by a continuous infusion of up to 40 mg/hour or fluid removal using renal replacement therapy. The investigators aimed for a net negative fluid balance of >1 ml/kg/hour (ideal body weight) with a target cumulative fluid balance of less than 1000 ml from the time of ICU admission. The recruitment rate was low, with only 23 included among 1144 patients who were screened. Despite the small sample size, a conspicuous reduction in the cumulative fluid balance was observed with active de-resuscitative measures at 5 days after randomization. The mean difference in the cumulative balance between groups was 5,814 ml (95% CI 2063 to 9565, P = .003). Although not powered to evaluate clinical outcomes, no adverse effects were noted with forced fluid removal in this pilot study.  

A combination of high PEEP, small volume resuscitation with 20% albumin, and diuretic treatment with furosemide (described as “PAL-treatment”) were evaluated among patients with acute lung injury by Cordemans et al., in a retrospective case-control study. The PEEP level (cm of H₂O) was set at the level of the intra-abdominal pressure (mm of Hg). This strategy resulted in a lower extravascular lung water index, intra-abdominal pressure, and cumulative fluid balance compared to control patients. The use of the PAL strategy resulted in a shorter duration of mechanical ventilation and ICU stay. The 28-day mortality was also lower with this approach. 

Another trail on the horizon is RADR-2 trial.

Our recommendations –
1. Remember “Four D’s” of fluid therapy (drug, duration, dosing and de-escalation). The analogy is similar to Antibiotic stewardship. There are four different phases in the time course of septic shock (resuscitation, optimization, stabilization and evacuation). Each phase requires a different therapeutic attitude regarding fluid administration.

2. Key strategies include – i. Minimisation of resuscitation fluid. ii. Early use of vasopressors. iii. Minimising prescription of maintenance fluids. iv. Use of diuretic therapy or renal replacement therapy.

2. Prevention is better than cure. So do not prescribe fluids routinely as maintenance to ICU patients.

3. Judicious use of diuretic therapy or renal replacement therapy to remove excess fluids.

4. Good clinical examination, bedside assessment of cumulative fluid balance and FiO2 requirements, CXR and biochemistry analysis will help you to switch from resuscitation phase to de-resuscitation phase.