Timing of CKRT Initiation and Discontinuation

Timing of CKRT Initiation and Discontinuation

Timing of initiation

Continuous kidney replacement therapy is initiated when a clinical indication is present, such as life-threatening disturbances of fluid, electrolyte or acid–base balance. The decision is based on the patient's complete clinical context rather than an isolated laboratory value [1].

Starting CKRT too late may lead to complications from acute kidney injury (AKI) and volume overload. Conversely, starting too early may expose patients who might not actually need kidney replacement therapy to its potential risks.

In STARRT-AKI, the accelerated initiation strategy did not reduce 90-day mortality compared with the standard strategy, while dependence on kidney replacement therapy increased among survivors. This finding supports avoiding routine accelerated initiation in the absence of an urgent indication; it does not justify delaying treatment when a clear indication exists [2].

In AKIKI and IDEAL-ICU, no consistent benefit of early initiation was demonstrated in the patients included in the two trials. Some signals of possible harm with accelerated initiation were observed, including:

— Impaired recovery of kidney function.

— Increased incidence of catheter-related bloodstream infections (Catheter-Related Bloodstream Infections).

— Higher rates of hypotension.

— And increased incidence of hypophosphatemia.

The more recent AKIKI-2 trial randomized patients to a standard initiation strategy versus a more-delayed strategy.

The more-delayed strategy (More delayed) produced a trend toward increased mortality, with an 11% increase in mortality within 60 days, P = 0.07.

Thus, accelerated initiation of kidney replacement therapy has no clear benefit, whereas AKIKI-2 suggests that patients may be harmed if treatment is delayed excessively beyond standard initiation strategies.

Trial criteria differ and are not used as a substitute for patient assessment [3–5].

Indications and reassessment of goals

For most patients, CKRT should be initiated in response to clear, concrete clinical indications, most commonly:

In patients who are appropriate candidates for escalation of care, initiating CKRT is also likely to be a reasonable option when oliguria persists for ≥48–72 hours.

Notably, overall mortality in these randomized trials, as in many other studies of critically ill patients with AKI requiring kidney replacement therapy, is approximately 50%.

In the absence of an indication for kidney replacement therapy involving an imminent threat to life, discussion of CKRT initiation should generally prompt both intensivists and nephrologists to reassess the patient's overall prognosis.

For many patients, it may be appropriate to discuss goals of care with them or their representatives before obtaining consent to initiate CKRT.

De-escalation and discontinuation or transition to another modality

Although randomized controlled trials are ongoing, there are currently no clinical trial data on which to base guidance for de-escalating or discontinuing CKRT.

Decisions to discontinue CKRT or transition to intermittent treatment generally depend on recovery of kidney function, urine output, hemodynamic stability and the patient's ability to tolerate the alternative modality. These indicators are used within the clinical assessment [1].

Observational studies suggest that the following criteria may be reasonable when considering discontinuation of kidney replacement therapy in patients with acute kidney injury, bearing in mind that no single urine output value determines discontinuation for all patients:

— Spontaneous urine output >500 mL/day

Or:

— Diuretic-enhanced urine output >2 liters/day.

Ongoing vasopressor requirements and a more positive cumulative fluid balance are also associated with intradialytic hypotension in patients transitioning from CKRT to intermittent hemodialysis (IHD).

In general, achieving hemodynamic stability without vasopressor support is a common criterion prompting consideration of transition from CKRT to IHD.

Experts also suggest correcting volume overload before discontinuing CKRT or transitioning to another kidney replacement modality.

CKRT dose

KDIGO recommends prescribing a dose that permits an actually delivered effluent volume dose of 20 to 25 mL/kg/hour in adults with AKI receiving CKRT [1].

Unlike intermittent hemodialysis (IHD), in which pre- and post-treatment urea measurements are used to calculate clearance and dose adequacy, continuous therapy (CKRT) uses the total effluent flow rate (Total Effluent Flow Rate, Qef) as a surrogate for clearance.

Qef represents the total clearance resulting from:

— Diffusion

— Convection

— And net ultrafiltration

Although some early trials suggested a potential benefit of higher doses, two subsequent large randomized trials, the VA/NIH ATN trial and the RENAL trial, showed that higher CKRT doses of 35–40 mL/kg/hour provided no benefit over lower doses of 20–25 mL/kg/hour.

Higher doses were instead associated with slightly higher rates of some complications, including:

— Hypophosphatemia.

— Hypotension.

(Delivered dose) of 20–25 mL/kg/hour.

This standard CKRT dose provides much slower instantaneous clearance than IHD, but a higher total daily or weekly solute removal dose than IHD three times weekly, and even than daily IHD.

The achieved dose may be lower than prescribed when treatment stops or because of filter clotting or procedures outside the unit. Actual treatment hours and the achieved effluent rate are therefore reviewed. When calculating Qef, the prescription components and UFnet should be defined in the same way on the device [1].

The difference between prescribed and actually delivered dose

In practice, several factors cause the actual achieved or delivered CKRT dose to be lower than prescribed, including:

— Loss of part of the effective surface area because of filter clotting or pore blockage

— The time required to change the filter

— The need for imaging examinations or procedures outside the intensive care unit.

— And other factors.

Observational studies suggest that actually delivered CKRT dose, rather than prescribed dose, is associated with clinical outcomes, although this relationship may be affected by confounding factors.

To overcome the usual difference between prescribed and delivered dose, some experts have recommended empirically increasing the prescribed dose by 20%–25%.

However, the actually delivered dose was slightly lower than prescribed in all groups of the ATN and RENAL trials.

For example, in the low-dose groups:

In ATN: 17.5 versus 20 mL/kg/hour

In RENAL: 22 versus 25 mL/kg/hour

Nevertheless, clinical outcomes were equivalent with actually delivered doses ranging from 17.5–22 mL/kg/hour.

It may therefore be unnecessary to increase the prescribed dose above 25 mL/kg/hour.

Can doses below 20 mL/kg/hour be used?

Although this remains controversial, it is theoretically possible that prescribed doses below 20 mL/kg/hour are no less effective than standard doses.

Trials are ongoing to investigate this, but no randomized controlled trial data currently support this strategy.

More comprehensive measures of solute removal adequacy may also be as important as prescribed Qef, including:

— Electrolyte homeostasis.

— Acid–base balance

— Potassium concentration

— pH

Potential risks of higher delivered CKRT doses include excessive or unwanted removal of:

— Electrolytes, particularly phosphate

— Micronutrients

— Medications

Effect of replacement solution administration site on dose

The site at which replacement solution is administered in the CKRT circuit affects dose, particularly in:

— CVVH

— CVVHDF

Post-filter replacement: Postfilter / Postdilution

When replacement solution is administered after the filter, in postdilution mode, solute removal is maximally efficient because the solute concentration in the ultrafiltrate equals its concentration in plasma water.

The trade-off, however, is that filtration of undiluted plasma water markedly increases the concentration of:

— Red blood cells.

— And proteins.

Along the filter, increasing the risk of filter clotting.

Filtration Fraction

The risk of clotting caused by hemoconcentration within a CKRT filter is traditionally estimated by calculating:

The filtration fraction (Filtration Fraction, FF)

This is the ratio of the total ultrafiltration rate to the plasma water flow rate.

To reduce clotting risk, FF should be kept below 20%–25%, meaning that less than one quarter of blood flow is filtered.

This can be achieved either by:

— Maintaining an adequate blood flow rate Qb that is 4 times higher than the filtration rate.

— Or relatively favoring pre-filter replacement solution.

Regional citrate anticoagulation (RCA) may be a possible exception, as clinical experience suggests it may permit higher FF values.

Pre-filter replacement: Predilution

When replacement solution is administered before the filter, in predilution mode, filter life is longer because plasma water is diluted before ultrafiltration, reducing FF as described above.

However, solute concentrations are diluted before blood enters the filter, causing some reduction in clearance.

The magnitude of this effect can be estimated by calculating the dilution factor.

Notably, FF is lowest with CVVHD.

Observational and quasi-randomized data also suggest that CVVHD may provide longer filter life than CVVH with pre-filter replacement solution.

Post-filter hematocrit as an alternative marker

Although FF is the traditional marker used to describe the effect of hemoconcentration on filter clotting risk, the following has recently been proposed:

Post-filter hematocrit as an alternative measure for predicting clotting risk.

However, data demonstrating superiority of either marker over the other remain limited.

Reassessing CKRT dose

Although delivering 20–25 mL/kg/hour is appropriate for most patients treated with CKRT in most circumstances, the dose should be reassessed frequently because critical illness changes dynamically over time.

In some circumstances, doses above 25 mL/kg/hour may become necessary.

Examples include:

— Rhabdomyolysis

— Tumor lysis syndrome

In these situations, the reduction in potassium clearance caused by AKI is exacerbated by the release of very large amounts of potassium into the circulation. CKRT doses exceeding 40 mL/kg/hour may be needed to maintain metabolic control.

Combining IHD and CKRT in severe hyperkalemia

In these situations, it may be necessary initially to use the high instantaneous clearance of IHD to correct life-threatening hyperkalemia rapidly, followed immediately by CKRT to provide a continuous, higher total daily clearance dose to maintain metabolic control.

When high CKRT doses are used, the dose should be reduced gradually toward the standard range once metabolic control is achieved and the underlying disease process improves.

The aim is to avoid excessive removal of:

— Phosphate

— Micronutrients

— Medications.

When is high-dose CKRT used in lactic acidosis?

One situation in which high-dose CKRT should not be used routinely, despite ongoing debate,

Is the treatment of lactic acidosis, particularly type A lactic acidosis,

That is, resulting from shock or organ-specific hypoperfusion.

Although lactate is a small, water-soluble molecule and can therefore be removed relatively easily by dialysis, with S ≈ 1,

Kidney replacement therapy generally has a very limited effect on serum lactate levels.

This is because the clearance provided by CKRT, even at very high doses, such as 50 mL/minute,

Remains extremely low compared with normal endogenous lactate clearance.

Endogenous lactate clearance in critically ill patients has been estimated at approximately 750–2,000 mL/minute.

CKRT therefore usually has only a small effect on serum lactate levels.

Changes in lactate levels in patients receiving CKRT can therefore generally be interpreted in much the same way as in patients not receiving CKRT.

Type B lactic acidosis

Although less common, kidney replacement therapy plays an important role in treating type B lactic acidosis caused by drug toxicity, such as metformin poisoning.

In this situation, however, intermittent hemodialysis (IHD) is often preferred to CKRT because it can achieve higher clearance.

Blood Flow

Compared with intermittent hemodialysis (IHD), the usual range of blood flow rates Qb used in continuous kidney replacement therapy (CKRT) is much lower.

As mentioned previously, Qb usually has very little effect on CKRT dose within the usual dose range.

At high CKRT doses, however, Qb may begin to affect clearance. In these situations, an adequate Qb relative to Qd or Qr,pre is needed to maintain highly efficient solute removal.

Otherwise, maintaining adequate Qb is necessary to prevent hemofilter clotting.

Although increasing Qb will always mathematically reduce the filtration fraction (Filtration Fraction, FF), the potential benefit for extending filter life begins to diminish with further increases in Qb, probably once it exceeds approximately 250 mL/minute.

For example, one trial randomized 100 patients receiving CKRT either without anticoagulation or with heparin to blood flow rates of 150 mL/minute or 250 mL/minute.

The study found no difference in filter life between the two groups.

This may be because increasing Qb in a circuit with fixed resistance also increases the risk of pressure alarms. These pressure alarms stop or slow the blood pump, which may promote filter clotting if they occur repeatedly.

Qb during regional citrate anticoagulation (RCA)

With regional citrate anticoagulation (Regional Citrate Anticoagulation, RCA), a lower blood flow rate, such as 120–150 mL/minute, is generally recommended.

There are two principal reasons:

— Lower Qb reduces the amount of citrate required to maintain adequate anticoagulation, thereby reducing the risk of metabolic complications associated with RCA.

— RCA is sufficiently effective to maintain filter patency even at relatively low Qb.

References

[1] KDIGO. Clinical Practice Guideline for Acute Kidney Injury. Kidney Int Suppl. 2012;2:1–138. https://kdigo.org/wp-content/uploads/2016/10/KDIGO-2012-AKI-Guideline-English.pdf.

[2] STARRT-AKI Investigators. Timing of Initiation of Renal-Replacement Therapy in Acute Kidney Injury. N Engl J Med. 2020;383:240–251. doi:10.1056/NEJMoa2000741.

[3] Gaudry S, et al. Initiation Strategies for Renal-Replacement Therapy in the Intensive Care Unit. N Engl J Med. 2016;375:122–133. doi:10.1056/NEJMoa1603017.

[4] Barbar SD, et al. Timing of Renal-Replacement Therapy in Patients with Acute Kidney Injury and Sepsis. N Engl J Med. 2018;379:1431–1442. doi:10.1056/NEJMoa1803213.

[5] Gaudry S, et al. Delayed versus more-delayed initiation of renal-replacement therapy for severe acute kidney injury (AKIKI 2). Lancet. 2021;397:1293–1300. doi:10.1016/S0140-6736(21)00350-0.

[6] Palevsky PM, et al. Intensity of Renal Support in Critically Ill Patients with Acute Kidney Injury. N Engl J Med. 2008;359:7–20. doi:10.1056/NEJMoa0802639.

[7] Bellomo R, et al. Intensity of Continuous Renal-Replacement Therapy in Critically Ill Patients. N Engl J Med. 2009;361:1627–1638. doi:10.1056/NEJMoa0902413.

[8] EXTRIP Workgroup. Extracorporeal Treatment for Metformin Poisoning: Systematic Review and Recommendations. Crit Care Med. 2015;43(8):1716–1730. doi:10.1097/CCM.0000000000001002.