Continuous veno-venous hemofiltration (CVVH) works by pumping blood from a large vein through a high-permeability hemofilter and returning it to the same circulation, removing plasma water and dissolved solutes by convection rather than by dialysis. A pump drives filtrate across the membrane in response to a transmembrane pressure gradient, and a balanced replacement fluid is infused to restore volume. The result is a slow, continuous, hemodynamically gentle clearance of fluid, urea, potassium, and middle molecules that the failing kidney can no longer manage.
The Convective Mechanism: Solvent Drag Across a Permeable Membrane
CVVH is a pure convection therapy. Blood enters the filter, and a pressure gradient pushes plasma water across a semipermeable membrane. Dissolved solutes travel with that water in a process called solvent drag: small molecules pass freely with the fluid, while larger proteins are retained. The membrane's effective pore size governs this sieving, and modern high-flux hemofilters pass molecules up to roughly 20,000 to 30,000 daltons.
The driving force is transmembrane pressure, the difference between hydrostatic pressure in the blood compartment and pressure on the filtrate side. Operators manipulate this gradient by adjusting blood flow, filtrate pump rate, and the pressure at which replacement fluid is delivered. Because the membrane is highly permeable to water, the filtrate rate is set primarily by the filtrate pump, not by the membrane's resistance. That distinction matters: in CVVH, the machine meters effluent flow directly, and transmembrane pressure adjusts to deliver it.
Clearance in convection depends on two variables: the ultrafiltration rate and the sieving coefficient of the solute. A solute with a sieving coefficient near 1.0 passes as freely as water. Urea, creatinine, and potassium sit near that value. Larger solutes and protein-bound drugs fall well below it. The practical implication is that convective clearance favors small and middle molecules, roughly up to the cutoff of the membrane, and is poor for highly protein-bound substances.
Replacement fluid restores what convection removes. It can be delivered before the filter (predilution) or after it (postdilution). Predilution dilutes blood entering the filter, lowering the fraction of plasma water removed and reducing filter clotting risk. Postdilution achieves higher clearance per liter of effluent because blood is not diluted before filtration, but it concentrates blood in the filter and raises clotting risk. This tradeoff drives most prescription decisions.
Quantified Context: Dosing, Flows, and the Numbers That Matter
CVVH prescriptions are written in milliliters per kilogram per hour of effluent. Typical continuous renal replacement therapy (CRRT) dosing falls in the range of 20 to 25 mL/kg/h for standard indications, with higher doses historically studied for sepsis. A 2009 multicenter trial found no survival benefit from higher-intensity dosing, which helped settle practice around moderate targets rather than aggressive ones. For a 70 kg patient, 25 mL/kg/h equals 1,750 mL/h of total effluent, a figure that anchors both clearance and fluid balance planning.
Blood flow typically runs 100 to 200 mL/min, with many programs targeting 150 to 200 mL/min. The filtration fraction, the ratio of effluent rate to plasma flow, is usually kept below roughly 20 to 25 percent in postdilution to limit hemoconcentration and filter failure. Replacement fluid rates often approximate 1,000 to 2,000 mL/h, split between pre- and postdilution depending on the anticoagulation strategy and clotting history.
Filter lifespan is a practical benchmark. Unplanned circuit clotting remains the leading cause of downtime, and many centers expect filters to last 24 to 72 hours with effective anticoagulation. Each circuit exchange costs staff time and interrupts clearance, so anticoagulation strategy directly affects delivered dose. The prescribed dose and the delivered dose diverge whenever the circuit is down.
Fluid balance is quantified separately from solute clearance. Net ultrafiltration, the difference between total effluent and replacement plus other inputs, is set to achieve a target weight change. Rates of 0 to 200 mL/h are common, and higher rates are used cautiously because rapid fluid removal can compromise cardiac output in unstable patients. The machine reports cumulative balance in liters, and hourly reassessment is standard in critical care.
Edge Cases: When the Standard Model Breaks Down
Anticoagulation is the first edge case. Regional citrate anticoagulation chelates calcium in the circuit, and the machine must return calcium to the patient systemically. Citrate accumulation can occur in severe liver dysfunction, producing a rising total-to-ionized calcium ratio that signals toxicity. Unfractionated heparin offers an alternative but carries bleeding and thrombocytopenia risks. No anticoagulation is used in patients with active bleeding, at the cost of markedly shorter filter life.
Protein-bound solute clearance is a second limit. Convection clears free drug and free solute, not the protein-bound fraction. Highly bound substances such as many antibiotics and several toxins are poorly removed, and drug dosing during CVVH must account for this. The same logic applies to endogenous protein-bound toxins, which is one reason convective therapy alone does not replicate all functions of a native kidney.
Volume status in shock is a third edge case. CVVH removes fluid, but in vasopressor-dependent patients the margin between adequate and excessive removal is narrow. Net ultrafiltration is often paused or minimized until hemodynamics stabilize. This is a clinical decision that balances clearance needs against perfusion risk, and it is why the continuous character of the therapy is valuable: the rate of removal can be throttled, unlike intermittent hemodialysis.
A counterintuitive detail: CVVH does not depend on a dialysate at all. In continuous veno-venous hemodialysis, dialysate flows countercurrent to blood and removes solute by diffusion, with clearance driven by concentration gradients. In CVVH, the entire clearance derives from the volume of filtrate generated. Adding dialysate converts the therapy to continuous veno-venous hemodiafiltration (CVVHDF), which combines convection and diffusion. Many modern machines offer all three modes, and mode selection shapes which solutes are cleared most efficiently.
Misconceptions: What CVVH Is Not
A persistent misconception is that CVVH "dials" the patient, meaning it corrects blood chemistry to normal targets. In practice, continuous therapy corrects gradually. The rate of solute removal is set by effluent flow, not by a dialysate prescription, and the approach to steady state is slower than in intermittent dialysis. Rapid correction of severe electrolyte derangement is a job for intermittent modalities, not typically CVVH.
Another misconception is that CVVH replaces kidney function comprehensively. It does not. The therapy removes fluid and small solutes, but it does not produce erythropoietin, activate vitamin D, or restore endocrine kidney functions. Clearance of middle molecules is partial, and clearance of protein-bound solutes is poor. Patients on CRRT remain dependent on the machine for the functions it can provide.
A third misconception treats higher dose as better. The evidence from large trials does not support intensifying dosing beyond standard targets for improved survival. Higher effluent rates increase cost, fluid turnover, and filter stress without demonstrated outcome benefit in the populations studied. Prescribing should match the indication, not chase a maximum.
Finally, CVVH is frequently described as "gentle" in an absolute sense. It is gentler than intermittent hemodialysis in hemodynamic terms because fluid removal and solute change happen slowly. It is not risk-free: circuit clotting, bleeding from anticoagulation, electrolyte shifts, and catheter complications are continuous hazards. Gentle relative to the alternative is not the same as benign.
Practical Actions: Putting the Prescription Into Practice
Clinicians set up CVVH around a small set of decisions, each with concrete thresholds.
Confirm the indication and target dose, commonly 20 to 25 mL/kg/h effluent, and calculate total effluent for the patient's weight. Select anticoagulation strategy with explicit monitoring: citrate with ionized calcium checks and total-to-ionized calcium ratio surveillance, or heparin with aPTT targets, or no anticoagulation for active bleeding. Set replacement fluid split mindful of filtration fraction, keeping postdilution filtration fraction under roughly 20 to 25 percent where clotting risk is high. Establish the fluid balance target as a net ultrafiltration rate, reassessing hourly against cumulative balance and hemodynamics. Plan drug dosing around the sieving behavior of convective clearance, since protein binding limits removal of many agents. Review circuit performance each shift: filter life, downtime, and delivered versus prescribed dose.
Red-flag warnings deserve separate emphasis.
Rising total-to-ionized calcium ratio on citrate anticoagulation suggests citrate accumulation and demands immediate review. A suddenly declining effluent rate or rising transmembrane pressure often signals filter clotting. Hypotension during net fluid removal requires reassessing the ultrafiltration rate, not pushing through it. Bleeding in a patient on systemic anticoagulation requires stopping the anticoagulant and weighing circuit downtime against hemorrhage. Catheter malfunction or malposition can drop blood flow and silently reduce delivered clearance.
These warnings are not exhaustive, and each institution has its own protocols for escalation.
Comparison: CVVH Against Alternative Modalities
| Modality | Primary clearance mechanism | Typical flow context | Best suited for |
|---|---|---|---|
| CVVH | Convection | Continuous, effluent-driven | Hemodynamically unstable patients needing fluid and middle-molecule removal |
| CVVHD | Diffusion | Continuous, dialysate countercurrent | Small-solute clearance with less replacement fluid burden |
| CVVHDF | Convection plus diffusion | Continuous, both fluids | Broad clearance when both mechanisms are desired |
| Intermittent hemodialysis | Diffusion | Intermittent sessions | Stable patients needing rapid, scheduled clearance |
| Peritoneal dialysis | Diffusion and convection | Continuous, peritoneal membrane | Selected patients without vascular access or with hemodynamic fragility |
The table summarizes mechanisms rather than outcomes, and modality choice depends on patient stability, access, and local capability. A given patient may move between modalities as physiology changes.
Timeline: What Happens During a CVVH Circuit
| Phase | What occurs | Practical significance |
|---|---|---|
| Prescription | Dose, anticoagulation, fluid targets set | Determines delivered clearance ceiling |
| Initiation | Circuit primed and connected to vascular access | Access quality sets achievable blood flow |
| Early hours | Replacement fluid and effluent flows stabilize | Steady-state approach begins |
| Steady state | Continuous solute and fluid removal near target | Delivered dose can be benchmarked |
| Circuit aging | Membrane fouling and clotting risk accumulate | Filter life commonly 24 to 72 hours |
| Circuit exchange | Clotting or scheduled change restarts the cycle | Downtime reduces delivered dose |
| Discontinuation | Therapy ends when indication resolves or modality changes | Transition planning matters |
The timeline reinforces a central point: CVVH is a continuous therapy, and its performance is measured over hours and days, not sessions. Downtime and interruptions are part of the dose delivered.
Limitations: What Continuous Hemofiltration Cannot Do
CVVH cannot replace a native kidney. It clears fluid and small solutes by convection, but its clearance of protein-bound and larger solutes is limited by membrane sieving, and its endocrine functions are absent. The therapy buys time and stability; it does not restore the organ.
CVVH also cannot be prescribed without reliable vascular access, adequate anticoagulation strategy, and trained staff. Circuit clotting is the dominant practical limitation, and every interruption in flow reduces the dose actually delivered. Programs that achieve good filter life also deliver more of the prescribed dose.
The evidence base has boundaries. Large trials have informed dosing, but many edge-case decisions, including precise anticoagulation targets and fluid removal rates in shock, rest on physiology and local protocol rather than definitive outcome trials. That is a limitation of certainty, not a reason to abandon the therapy. CVVH remains an indispensable tool for critically ill patients with acute kidney injury, and understanding its convective core helps clinicians prescribe it honestly.