Use in Continuous Renal Replacement Therapy

SUMMARY

• Deep et al (2024)¹ conducted a retrospective, single-center study (2010-2019) evaluating the use of epoprostenol as a sole anticoagulant during continuous renal replacement therapy (CRRT) in critically ill pediatric patients with liver disease (N=101).
  • Patients receiving epoprostenol anticoagulation experienced a median filter life of 48 hours (interquartile range [IQR], 32-72), with comparable outcomes between those with acute liver failure (ALF) and those with other liver diseases (ALF vs other cases, 50 [IQR, 28.75-72] vs 48 [IQR, 34-70] hours; P=0.839).
  • Major bleeding was observed in 21/353 (5.9%) filters (1.13 episodes/1000 CRRT hours) and minor bleeding was observed in 18/353 (5.1%) filters (0.97 episodes/1000 CRRT hours).
• Sondhi et al (2024)² conducted a retrospective chart review at a quaternary pediatric hospital (2017-2022) to compare the efficacy, safety, and cost-effectiveness of epoprostenol and heparin or citrate as anticoagulants in patients <18 years of age who underwent CRRT (N=101).
  • In cases of unplanned first filter changes due to clogging or clotting, patients receiving epoprostenol experienced a longer median (IQR) time until filter replacement (29 hours [IQR, 27-48]) vs patients receiving citrate (21 hours [IQR, 12-25]) or heparin (18 hours [IQR, 15-27]).
  • The median ratio of filters used to days on therapy was lower with epoprostenol (0.53 [IQR, 0.42-0.85]) vs citrate (1 [IQR, 0.50-1.13]) or heparin (0.75 [IQR, 0.50-1.08]).
  • For all 3 groups, there was no difference in the median platelet count on day 1 to day 7 of treatment.
• Aldairi et al (2024)³ conducted a systematic review and meta-analysis to compare the efficacy of prostacyclin-based anticoagulation (PgI₂ alone or combined) and other strategies (heparin, citrate, or none) in critically ill children and adults requiring extracorporeal support, focusing on the circuit lifespan.
  • Among 11 studies, the reported mean circuit lifespan was 29.7 hours in patients receiving a prostacyclin-based anticoagulation series (95% confidence interval [CI], 21.8-37.6; 974 circuits) and 27.2 hours in patients receiving a heparin- or citrate-based anticoagulation series (95% CI, 19.4-34.1; 3029 circuits).
  • Prostacyclin-based vs heparin- or citrate-based strategies were associated with a significantly lower bleeding rate (9.5% vs 17.1%; P<0.001) but showed no significant difference in thrombotic events.
• Gainza et al (2006)⁴ conducted a retrospective study to evaluate the efficacy, safety, and cost-effectiveness of epoprostenol as an alternative anticoagulant to heparin in patients undergoing CRRT, particularly in cases where heparin is contraindicated due to hypercoagulability, thrombocytopenia, or recent hemorrhage (N=38).
  • Patients were categorized into the following 2 groups: group 1, received epoprostenol as a first-line anticoagulant (n=10); group 2, received epoprostenol as a second-line anticoagulant after heparin or no anticoagulation (n=28).
  • The mean filter duration in patients was 38±13 hours during epoprostenol administration.
  • In group 2, the mean±standard deviation (SD) filter duration increased from 23±12 hours (pre-epoprostenol) to 38.2±11.9 hours (post-epoprostenol) (P=0.0001).
• Fiaccadori et al (2002)⁵ conducted a retrospective, observational study to evaluate the safety and efficacy of epoprostenol as the sole antihemostatic agent during continuous veno-venous hemofiltration (CVVH) in critically ill patients with acute renal failure (ARF; N=51).
  • The mean APACHE II score was 27.2±7.8. Acute tubular necrosis was identified in 44 patients (86.3%), and 14 patients (21%) required mechanical ventilation. The in-hospital mortality rate was 54% (28 patients).
  • The median duration of the circuit was 15.0 hours (95% CI, 13.0-16.5). The mean time to failure was 19.4 hours (95% CI, 16.7-22.2).
  • Mean urea clearance was 22.8 mL/min, with sieving coefficients for urea mostly above 0.85; this indicated that from the 8th to the 12th hour, most fibers in the filter were patent.
  • Four patients (7.8%) experienced major bleeding (upper gastrointestinal tract hemorrhage) during CVVH (1.0 episode per 1000 patient hours of treatment; 95% CI, 0.42.6).
  • Hypotension episodes requiring therapeutic intervention occurred in 15.5% of CVVH sessions (95% CI, 9.2-25.3) but did not require interruption of epoprostenol administration.
• Additional studies describing simultaneous administration of VELETRI as an anticoagulant during CRRT have been referenced for your review.^6-10

CLINICAL DATA

Deep et al (2024)¹ conducted a retrospective, single-center review of clinical practice (2010-2019) to describe the use of epoprostenol as a sole anticoagulant during CRRT in critically ill children with liver disease (N=101).

Study Design/Methods

• The review included pediatric patients with liver disease (ie, ALF, post-liver transplant, chronic liver disease, and other diagnoses) who received epoprostenol as the sole anticoagulant during at least 1 filter session of CRRT.
• A total of 101 patients with liver disease were treated with CRRT between 2010 and 2019; 96 (95%) of these were included in the analysis. Across 108 patient admissions (ALF, n=58; other liver disease, n=50), patients were treated with epoprostenol using 353 CRRT filters, totaling 18,508 hours of therapy.

Results

• The median age at CRRT initiation was 5 years (IQR, 0.5-13), and the median weight was 18.4 kg (IQR, 6.8-40.3).
• The median effective total duration of CRRT per admission was 96 hours (IQR, 56-211) for patients with ALF and 139 hours (IQR, 80-274) for those with other liver diseases (P=0.292).
• In patients receiving epoprostenol, the median filter life was 48 hours (IQR, 32-72), with comparable outcomes between patients with ALF and those with other liver diseases (ALF vs other cases, 50 [IQR, 28.75-72] vs 48 [IQR, 34-70] hours; P=0.839).
• The effective 60-hour filter-survival rate was 60.5%. The median life of the first filter for each patient admission was 48 hours (IQR, 31.5-72), which was similar to the overall median filter duration.
• Serum creatinine and urea levels significantly improved by 48 hours following CRRT initiation (P <0.001). The median serum creatinine level was 61 µmol/L (IQR, 36-146) at the start of CRRT vs 41.5 µmol/L (IQR, 26.5-73.5) at 48 hours, and the median urea level was 6.5 mmol/L (IQR, 3.2-18.1) at the start of CRRT vs 4.8 mmol/L (2.3-7.5) at 48 hours.
• Major bleeding was reported in 21/353 (5.9%) filters (1.13 episodes/1000 CRRT hours), and minor bleeding was reported in 18/353 (5.1%) filters (0.97 episodes/1000 CRRT hours). Hypotension was reported in 41/353 (11.6%) filters (2.22 episodes/1000 CRRT hours), with no significant change in the mean arterial pressure at 1 hour after initiation.
• Among filters used in patients with an initial platelet count ≤50×10⁹/L, the rates of major bleeding, minor bleeding, and hypotension were 0.74, 0.49, and 2.23 per 1000 hours, respectively.

Sondhi et al (2024)² conducted a retrospective chart review at a quaternary pediatric hospital (2017-2022) to compare the efficacy, safety, and cost-effectiveness of epoprostenol and heparin or citrate as anticoagulants in patients <18 years of age who underwent CRRT (N=101).

Study Design/Methods

• Among the 101 patients evaluated, 44 (43.6%) received epoprostenol, 38 (37.6%) received heparin, and 19 (18.8%) received citrate.
• Patients receiving epoprostenol were associated with a higher proportion of planned first filter changes (43%) vs those receiving citrate (11%) or heparin (29%).

Results

• In cases of unplanned first filter changes due to clogging or clotting, patients receiving epoprostenol experienced a longer median time until filter replacement (29 hours [IQR, 27-48]) vs those receiving citrate (21 hours [IQR, 12-25]) or heparin (18 hours [IQR, 15-27]).
• The median ratio of filters used to days on therapy was lower with epoprostenol (0.53 [IQR, 0.42-0.85]) vs citrate (1 [IQR, 0.50-1.13]) or heparin (0.75 [IQR, 0.50-1.08]).
• The median filter duration was the longest with epoprostenol (6 days [IQR, 3-14]), followed by heparin (5 days [IQR, 2-15]) and citrate (3 days [IQR, 2-12]).
• Patients receiving epoprostenol showed an increase in the median platelet count from 6 hours prior to initiating CRRT (54 [IQR, 35.5-106]) to 24 hours after initiating CRRT (60 [IQR, 55-131]) (P<0.001).
• Patients receiving heparin showed a decline in the median platelet count from 6 hours prior to initiating CRRT (103 [IQR 55-131]) to 24 hours after initiating CRRT (77 [IQR 44-107]) (P=0.020).
  • Patients receiving citrate showed no difference in the platelet count at any time point.
  • For all 3 groups, there was no difference in the median platelet count for day 1 to day 7 of treatment.
• Major bleeding events were reported in 2 patients (5.3%) treated with heparin and 1 patient (2.3%) treated with epoprostenol.

Aldairi et al (2023)³ conducted a systematic review and meta-analysis that compared the efficacy of PgI₂ alone or in association with another anticoagulant and other strategies (heparin, citrate, or none) in critically ill children and adults requiring extracorporeal support, focusing on circuit lifespan.

Study Design/Methods

• A total of 2148 references were identified; 17 studies (13 observational studies and 4 randomized controlled trials) met the eligibility criteria; these studies enrolled 1333 patients, with 179 (13.4%) children and 1154 (86.6%) adults.
• Sixteen studies included 1261 patients undergoing CRRT, and 1 study included 72 patients treated with a molecular absorbent recirculation system.

Results

• Among 11 studies, the reported mean circuit lifespan was 29.7 hours in patients receiving a prostacyclin-based anticoagulation series (95% CI, 21.8-37.6; 974 circuits) and 27.2 hours in patients receiving a heparin- or citrate-based anticoagulation series (95% CI, 19.4-34.1; 3029 circuits).
  • No significant difference was observed in the circuit lifespan between prostacyclin-based and heparin- or citrate-based anticoagulation (mean difference, 2.5 hours; 95% CI, -12.0 to 16.9; P=0.74; 4003 circuits).
• Bleeding was observed in 9.5% of patients in the prostacyclin-based anticoagulation group vs 17.1% of patients in the control group; this difference was statistically significant (logOR, -1.14; 95% CI, -1.91 to -0.37; P <0.001).
• Prostacyclin-based vs heparin- or citrate-based strategies were associated with a significantly lower bleeding rate (9.5% vs 17.1%; P<0.001) but showed no significant difference in thrombotic events.
• Thrombotic events were reported in 3.6% of patients receiving prostacyclin-based anticoagulation therapy and 2.2% of patients receiving heparin or citrate.

Gainza et al (2006)⁴ conducted a retrospective study to evaluate the efficacy, safety, and cost-effectiveness of epoprostenol as an alternative anticoagulant to heparin in patients undergoing CRRT, particularly in cases where heparin is contraindicated due to hypercoagulability, thrombocytopenia, or recent hemorrhage.

Study Design/Methods

• A total of 248 patients were treated with CRRT, of which 38 (15%) patients received epoprostenol for >72 hours. These patients were categorized into the following 2 groups: group 1, received epoprostenol as a first-line anticoagulant (n=10); group 2, received epoprostenol as a second-line anticoagulant after heparin or no anticoagulation (n=28).

Results

• The mean±SD filter duration in patients was 38±13 hours during epoprostenol administration.
• In group 2, the mean±SD filter duration increased from 23±12 hours (pre-epoprostenol) to 38.2±11.9 hours (post-epoprostenol) (P=0.0001).
• In patients with filter hypercoagulability, the mean±SD filter duration improved from 19.9±10.2 hours to 35.2±9.9 hours (P=0.0029).
• The filter consumption rate prior to epoprostenol administration was 1.5±1.3 filters/day and after epoprostenol administration was 0.7±0.3 filters/day (P=0.0022).
• In patients from group 2 receiving both heparin and epoprostenol simultaneously (n=6), the filter duration improved from 14.8±8.9 hours to 31.6±10.3 hours (P=0.07).
• Hemorrhage occurred in 7 patients (18%); 5 patients required transfusion and drug discontinuation.
• Hypotension occurred in 7 patients (18%) and resolved within 24 hours without discontinuing therapy.
• No deaths were attributed to epoprostenol.

Fiaccadori et al (2002)⁵ conducted a retrospective, observational study to evaluate the safety and efficacy of epoprostenol as the sole antihemostatic agent during CVVH in critically ill patients with ARF.

Study Design/Methods

• The study included 51 consecutive patients with ARF treated with bicarbonate-based CVVH using prostacyclin at 4 ng/kg/min (infused prefilter).

Results

• The mean±SD APACHE II score was 27.2±7.8. Acute tubular necrosis was identified in 44 patients (86.3%), and 14 patients (27.4%) required mechanical ventilation. The in-hospital mortality rate was 54% (n=28).
• The median time to failure was 15.0 hours (95% CI, 13.0-16.5). The mean time to failure was 19.4 hours (95% CI, 16.7-22.2).
• Mean urea clearance was 22.8 mL/min, with sieving coefficients for urea mostly above 0.85; this indicated that from the 8th to the 12th hour, most fibers in the filter were patent.
• During the first 3 days of CVVH, there was a slight decrease in the platelet count, which was not statistically significant (P=0.25). The mean daily percentage reduction during the entire treatment period was -3.2% (95% CI, -8.3 to 2.2; P=0.24).
• Four patients (7.8%) experienced clinically important major bleeding (upper gastrointestinal tract hemorrhage) during CVVH (1.0 episode per 1000 patient hours of treatment; 95% CI, 0.4-2.6).
• Minor bleeding events were reported in 2 (3.9%) patients (1 event of sternal wound bleeding and 1 event of tracheostomy bleeding).
• No deaths were attributed to hemorrhage.
• Hypotension episodes requiring therapeutic intervention occurred in 15.5% of CVVH sessions (95% CI, 9.2-25.3) but did not require interruption of epoprostenol administration.

Literature Search

A literature search of MEDLINE^®, Embase^®, BIOSIS Previews^®, and Derwent Drug File (and/or other resources, including internal/external databases) pertaining to this topic was conducted on 01 October 2025.