Citrate Toxicity in CRRT: Risks, Identification, Management

Citrate toxicity during continuous renal replacement therapy (CRRT) refers to a cluster of metabolic disturbances that arise when the body cannot break down citrate fast enough to keep pace with the rate it is being infused as an anticoagulant. The hallmark signs are a falling ionized calcium level, a widening gap between total and ionized calcium, and metabolic derangements that can be difficult to untangle from the patient’s underlying illness. Though serious, citrate accumulation is identifiable through careful monitoring and is manageable in most cases without stopping therapy altogether.

Why Citrate Is Used in CRRT

Citrate serves a straightforward purpose in the CRRT circuit: it binds ionized calcium in the blood as it leaves the patient, and without free ionized calcium, the clotting cascade stalls. This keeps the filter from clotting and extends its useful life. The calcium-citrate complexes that form are partly removed by the filter itself and partly returned to the patient’s bloodstream. Once back in the body, citrate is metabolized primarily by the liver, and to a lesser extent by the kidneys and skeletal muscle, into bicarbonate. Each molecule of citrate can yield three molecules of bicarbonate, and the calcium that was bound gets released back into circulation.¹ A separate calcium infusion is given through a central line to replace what the circuit removes, keeping the patient’s systemic ionized calcium in a safe range.

This approach has a major safety advantage over heparin-based anticoagulation: citrate’s anticoagulant effect is confined to the circuit. It does not thin the blood systemically, which matters enormously for critically ill patients who are already at high risk for bleeding. A large randomized trial found that bleeding complications dropped from about 17% with systemic heparin to roughly 5% with regional citrate anticoagulation.² A meta-analysis of randomized trials confirmed a roughly two-thirds reduction in bleeding risk with citrate compared to systemic heparin.³ The tradeoff is that citrate demands closer metabolic surveillance than heparin does.

What Happens When Citrate Accumulates

When the liver and other organs cannot metabolize citrate quickly enough, unprocessed citrate builds up in the bloodstream. Because citrate binds ionized calcium, a rising citrate level drives ionized calcium down. The body responds by mobilizing calcium stores, and clinicians respond by infusing more calcium, so total calcium in the blood can actually climb. This creates a characteristic and dangerous pattern: the total calcium looks normal or even elevated while the ionized calcium, the fraction that actually matters for heart rhythm and muscle function, drops to dangerously low levels.

One early case report captured this vividly. During CRRT with regional citrate anticoagulation, a patient’s ionized calcium fell as low as 2.72 mg/dL (the normal range is roughly 4.5 to 5.6 mg/dL), while total calcium climbed as high as 15 mg/dL, producing a total-to-ionized calcium ratio of 3.5 to 1.⁴ That widening ratio is the clearest laboratory fingerprint of citrate accumulation and the basis for the monitoring strategies used today.

Severe ionized hypocalcemia can cause cardiac arrhythmias, hypotension, muscle spasm, and in extreme cases cardiac arrest. But the metabolic picture can get more complicated than just low calcium. Unmetabolized citrate is an unmeasured anion, so it can widen the anion gap. And because citrate that is metabolized produces bicarbonate, patients can swing between acidosis (when citrate is accumulating and not being converted) and alkalosis (when a large citrate load is suddenly metabolized). One study of patients with severe liver dysfunction on long-term CRRT found metabolic alkalosis with a pH above 7.5 in about a quarter of cases.⁵

Who Faces the Highest Risk

Because the liver is the primary site of citrate metabolism, liver failure is the risk factor that gets the most attention. The KDIGO clinical practice guidelines for acute kidney injury have historically listed severe liver failure as a major contraindication to regional citrate anticoagulation due to the risk of citrate accumulation.⁶ The concern is straightforward: if the organ responsible for clearing citrate is badly damaged, infusing citrate continuously is risky.

In practice, though, the relationship between liver injury and citrate accumulation turns out to be less predictable than guidelines suggest. A study of patients with severe liver dysfunction found citrate accumulation (defined as a total-to-ionized calcium ratio above 2.4) in about 23% of patients, yet found no clear correlation between the degree of citrate accumulation and standard liver function test results.⁵ That unexpected finding suggests liver function tests alone do not reliably predict who will accumulate citrate. Other conditions traditionally flagged as risk factors include severe shock with lactic acidosis and profound hypoxemia, both of which impair the cellular machinery needed to metabolize citrate.⁷

A recent large study looked specifically at whether common markers of impaired tissue perfusion predicted citrate accumulation and found that surrogate markers of poor global perfusion did not show a statistically significant association with accumulation.⁸ This complicates the clinical picture: the patients most likely to accumulate citrate are not always the ones you would flag based on standard hemodynamic or laboratory markers. Vigilant monitoring matters more than trying to predict trouble from a checklist.

Recognizing Citrate Accumulation at the Bedside

The most widely used surrogate for citrate accumulation is the ratio of total calcium to ionized calcium. A ratio above 2.5 is the threshold most commonly cited in the literature, though some centers use 2.4 or even 2.25. A rising ratio signals that citrate-bound calcium is piling up in the blood faster than the body can process it. In a pediatric study of liver failure patients on CRRT, 70% experienced at least one episode where the ratio exceeded 2.5 for more than 48 hours, though frank clinical deterioration requiring treatment interruption was rare.⁹

Other laboratory values have been proposed as alternatives or supplements: pH, anion gap, base excess, and the strong ion gap. A head-to-head comparison of these surrogate markers found that the uncorrected calcium ratio had the best correlation with actual measured citrate levels, with a moderate fit. Every other marker tested, including pH, anion gap, albumin-corrected anion gap, standard base excess, and strong ion gap, correlated poorly.¹⁰ Even the calcium ratio, though the best available bedside tool, explained only about 40% of the variance in citrate levels. There is no perfect bedside marker, which is why protocols rely on frequent, repeated measurements rather than any single number.

On the circuit side, post-filter ionized calcium is monitored to ensure the citrate dose is adequate to prevent filter clotting. A target of roughly 0.25 to 0.35 mmol/L is standard. But one trial found that maintaining a fixed citrate concentration of 4 mmol/L achieved acceptable filter lifespans of around 60 hours even without adjusting based on post-filter calcium readings, and both groups had similar rates of filter clotting.¹¹ These results suggest that the traditional approach of chasing a tight post-filter target may be more labor-intensive than necessary, though larger confirmatory studies are still needed before protocols shift broadly.

Managing Citrate-Related Complications

When citrate accumulation is detected, clinicians have several tools before resorting to stopping citrate entirely. The most common interventions are reducing the citrate infusion rate, increasing the dialysis or filtration clearance to remove more citrate across the filter, adjusting the ultrafiltration rate, and increasing the calcium replacement infusion. In the pediatric liver failure study mentioned earlier, these adjustments were sufficient in nearly all patients; treatment was interrupted in only two patients, and for fewer than six hours each.⁹ This pattern holds in adult data as well: complete discontinuation of citrate is uncommon when accumulation is caught early.

Calcium replacement is the most immediate intervention for symptomatic ionized hypocalcemia. Both calcium gluconate and calcium chloride are used, with the choice depending on institutional protocol and the clinical situation. Calcium chloride delivers more elemental calcium per milliliter, but calcium gluconate is often preferred for peripheral infusion because it is less caustic to veins. Regardless of the formulation, the key is continuous titration based on frequent ionized calcium checks rather than a fixed drip rate.

There is growing interest in protocol-driven approaches that minimize the frequency of manual adjustments. A prospective study of an algorithm designed to maintain stable blood calcium found that a structured protocol reduced the incidence of both citrate accumulation and calcium disturbances, cutting down the need for rate changes to the citrate and calcium infusions during therapy.¹² Similarly, a comparison of fixed-rate versus variable-rate citrate protocols found that fixed-rate dosing was associated with fewer episodes of hypocalcemia and less total calcium supplementation.¹³ The direction of the field is toward simplifying protocols in a way that reduces both nursing workload and metabolic swings.

Is “Citrate Toxicity” Even the Right Term

There is an active debate about whether the phrase “citrate toxicity” is misleading. A recent review argued that the term should be abandoned entirely, on the grounds that citrate itself is not toxic in any pharmacological sense. The metabolic complications attributed to citrate, especially hypocalcemia, are really consequences of calcium chelation and impaired metabolism, not direct toxic effects. The same review pointed out that citrate is well known to cause alkalinization, not acidosis, and questioned the longstanding assumption that citrate accumulation causes metabolic acidosis.¹⁴

This is more than semantic quibbling. When clinicians hear “toxicity,” they think of a substance that is inherently harmful at a given dose, which can lead to unnecessary avoidance of citrate anticoagulation in patients who would benefit from it. A survey of physicians in the United States found that the perceived risk of hypocalcemia (cited by 52% of respondents) and general citrate safety concerns (42%) were the leading reasons clinicians chose to use no anticoagulant at all in heparin-intolerant patients, even though running CRRT without anticoagulation shortens filter life and increases circuit downtime.¹⁵ Those same respondents estimated that about 37% of patients on citrate develop hypocalcemia and about 17% develop “citrate lock,” though these estimates likely overstate the problem compared to what protocol-driven programs report.

Reframing the conversation around “citrate accumulation” rather than “citrate toxicity” may help align clinical language with the actual pathophysiology: the problem is impaired clearance and its downstream effects on calcium and acid-base balance, not a poison in the bloodstream.

How the CRRT Mode Affects Citrate Clearance

CRRT can be delivered in several modes, and the choice of mode turns out to matter for citrate handling. The two most common are CVVHD (continuous venovenous hemodialysis, which relies on diffusion) and CVVHDF (continuous venovenous hemodiafiltration, which combines diffusion with convection). A recent study found that CVVHD carried a significantly higher risk of citrate accumulation than CVVHDF, with a hazard ratio of about 2.3.¹⁶ The cumulative risk of accumulation was substantially higher in the CVVHD group throughout the treatment course.

This finding has a plausible mechanistic explanation. Citrate is a small molecule, and its clearance across the hemofilter depends on the combination of solute removal mechanisms available. When convection is added to diffusion, the filter removes more citrate per unit time, leaving less to return to the patient. In pediatric patients, citrate clearance across the filter was measured at roughly 31 to 38 mL/min per 1.73 m², similar to urea clearance, and remained stable whether clearance was achieved purely by convection or split between convection and diffusion.¹⁷ What differs between modes is the total volume of fluid exchange and how efficiently the circuit removes citrate before it reaches the patient. For patients at higher risk of accumulation, choosing CVVHDF over pure CVVHD may offer an additional margin of safety, though this consideration has to be weighed against other clinical factors driving mode selection.

Citrate Versus Heparin in Broader Safety Terms

Despite the concerns about citrate accumulation, the overall safety profile of citrate anticoagulation compares favorably to heparin. The large randomized trial comparing the two found that citrate cut bleeding complications by roughly two-thirds compared to systemic heparin, though it did not reduce the proportion of patients needing blood transfusions.² A smaller trial similarly found bleeding rates of about 12% with citrate versus 42% with heparin.¹⁸

One finding from the large trial that deserves attention is that the citrate group had a higher rate of new infections compared to the heparin group (68% versus 55%).² The reason for this association is unclear. It may relate to the immunomodulatory effects of citrate, or it may reflect differences in calcium handling that affect immune function, or it may be a statistical artifact of a single trial. Regardless, it is a signal worth tracking in future research.

The practical takeaway for clinicians weighing anticoagulation options is that citrate accumulation, while real and requiring attention, is a manageable complication in a technique that substantially reduces the most feared complication of circuit anticoagulation: bleeding in patients who are already critically ill. The risk is not zero, but the evidence consistently shows that for most patients, the metabolic vigilance citrate demands is a better tradeoff than the bleeding risk that heparin carries.