Are there alternative anticoagulation strategies to heparin or citrate for intraoperative cell salvage?

Comment by InpharmD Researcher

The only studied alternatives to acid-citrate dextrose solution A (ACD-A) anticoagulant for plasmapheresis or cell salvage are heparins (primarily unfractionated heparin). While citrate is generally preferred by physicians in plasma exchange (due to the risk of HIT), emerging evidence from Europe suggests heparinized saline has fewer adverse events than ACD-A for cell saving (Table 3). One ex vivo study suggests argatroban may be an alternative anticoagulant for blood salvage in patients with HIT, but the optimal dose is unknown.

Background

Guidelines from the United Kingdom Association of Anaesthetists on perioperative cell salvage recommend using either heparinized saline or acid-citrate dextrose (ACD) as the anticoagulant. The red cells are washed using intravenous saline 0.9% and then pumped into a bag for re-infusion to the patient. While there are no absolute contraindications to cell salvage, a history of heparin-induced thrombocytopenia (HIT) is a contraindication to using heparin as the anticoagulant; an anticoagulant solution containing acid-citrate dextrose should be used instead. If heparinized saline is used as the anticoagulant solution, care must be taken to add the correct volume and concentration of heparin and label the bag clearly so that it is not accidentally given intravenously. [1]

A 2018 review discussed the use of various anticoagulants in the setting of plasma exchange, including clinical considerations for optimal management of these patients. The review was compiled based on discussion by 10 physicians from multiple institutions, with consensus data presented. Primarily, citrate or heparin are considered to prevent clotting in the extracorporeal circuit, with citrate generally preferred, but heparin is reserved as a secondary option due to the potential for clinically significant bleeding and risk of HIT. [2]

A 2023 case series described 11 patients who experienced severe hypotension following reinfusion of autologous blood processed with a cell saver and anticoagulated with acid-citrate-dextrose solution A (ACD-A) during off-pump coronary artery bypass surgery. These hypotensive episodes, distinct from those typically associated with bypass surgery, were immediate, necessitating the use of vasopressors or inotropic agents; two cases required cardiopulmonary resuscitation. The events were characterized by significant decreases in cardiac output and mixed venous oxygen saturation. A retrospective root cause and prospective healthcare failure mode and effect analysis failed to conclusively identify the underlying mechanism, though incomplete removal of ACD-A from salvaged blood was proposed as a plausible contributor. Patient characteristics and interventions were meticulously documented, revealing that all patients received ACD-A as the anticoagulant during cell salvage. Reinfusion volumes varied, with no leukocyte depletion filters used. Among the 11 cases, hypotension was associated with nadir mean blood pressures as low as 23 mmHg, with mixed venous oxygen saturations dropping to as low as 38%. Transitioning to a heparin-based anticoagulant for cell salvage during a subsequent 1.5-year period involving 513 cardiac procedures eliminated further occurrences of such complications. These observations align with reports from the United Kingdom’s "Severe Hazards of Transfusion," which documented 31 cases of hypotension after reinfusion of salvaged blood since 2010, 22 of which involved ACD-A. [3]

A 2016 ex-vivo study investigated the use of argatroban as an anticoagulant in a blood salvage system, comparing its efficacy and residual concentrations to unfractionated heparin (UFH). Blood from 23 surgical patients contraindicated for blood salvage use was processed using the Continuous-Auto-Transfusion-System (C.A.T.S.®; Fresenius Kabi) with anticoagulation provided by either 5 mg, 50 mg, or 250 mg of argatroban in 1000 mL saline, or 25,000 U of heparin. Patency of the system and the extent of anticoagulant removal through emergency and high-quality washing protocols were evaluated. In 3 of 8 (38%) cases using 5 mg of argatroban, clotting occurred, rendering the system ineffective. High-quality washing removed 89-95% of argatroban, whereas emergency washing resulted in a lower clearance of 60-90%. Clearance was higher in the group with 50 mg argatroban compared to 5 mg or 250 mg, but the relative clearance among the groups did not differ significantly. Heparin removal was inconsistent, with residual anti-Xa activity detected in 20% of samples. Residual argatroban concentrations ranged from 55 ng/mL to 6810 ng/mL, increasing proportionally with the initial dose. Using an initial concentration of 50 mcg/mL resulted in a red cell concentrate containing approximately 0.794 mcg/mL of argatroban after high-quality wash, which could lead to a plasma concentration of 132 ng/mL upon reinfusion (argatroban steady state reference values: 300-600 ng/mL). The 250 mg argatroban dose generated significantly higher residual levels, posing a risk of unintended anticoagulation upon transfusion. Hematocrit levels were comparable across groups both before surgery and in the reservoir bag. After washing, hematocrits in re-transfusion bags were about double the initial patient levels and were similar across both washing methods. Based on these findings, anticoagulation with 50 mg of argatroban maintained system patency while being reduced significantly during washing, though additional dose-finding studies are necessary to identify the optimal concentration for clinical use. Argatroban 50 mg was selected due to similar anticoagulant potency to heparin 25,000 U, with the other two argatroban doses attempting to find lower and upper dosing ranges. Argatroban 5 mg concentrations were inadequate due to clotting issues, indicating it was too low to maintain system patency. A concentration of 50 mcg/mL in the solution, after washing, resulted in a residual argatroban concentration that could be safely re-infused without causing significant anticoagulation. However, emergency washing yielded higher residual concentrations, potentially leading to full anticoagulation and posing bleeding risks. The highest concentration tested, 250 mcg/ml, resulted in excessive anticoagulation, especially after emergency washing, suggesting it should be avoided. The study concludes there is a need for further research to identify safe argatroban concentrations below 50 mcg/mL for anticoagulating salvaged blood. [4]

References:

[1] Klein AA, Bailey CR, Charlton AJ, et al. Association of Anaesthetists guidelines: cell salvage for peri-operative blood conservation 2018. Anaesthesia. 2018;73(9):1141-1150. doi:10.1111/anae.14331
[2] Shunkwiler SM, Pham HP, Wool G, et al. The management of anticoagulation in patients undergoing therapeutic plasma exchange: A concise review. J Clin Apher. 2018;33(3):371-379. doi:10.1002/jca.21592
[3] Beersemans M, Rex S, Degezelle K, et al. Severe Hypotension After Reinfusion of Autologous Blood Processed With a Cell Saver and Anticoagulated With Acid-Citrate-Dextrose Solution A. J Cardiothorac Vasc Anesth. 2023;37(11):2397-2399. doi:10.1053/j.jvca.2023.07.034
[4] Beiderlinden M, Brau C, di Grazia S, Wehmeier M, Treschan TA. Argatroban for anticoagulation of a blood salvage system - an ex-vivo study. BMC Anesthesiol. 2016;16(1):37. Published 2016 Jul 15. doi:10.1186/s12871-016-0204-3

Literature Review

A search of the published medical literature revealed 4 studies investigating the researchable question:

Are there alternative anticoagulation strategies to heparin or citrate for intraoperative cell salvage?

Level of evidence

C - Multiple studies with limitations or conflicting results  Read more→



Please see Tables 1-4 for your response.


 

Regional citrate anticoagulation vs systemic heparin anticoagulation for double-filtration plasmapheresis

Design

Retrospective cohort study

N= 23 patients

Objective

To compare the efficacy and safety of regional citrate anticoagulation (RCA) with systemic heparin (Hep) anticoagulation during double-filtration plasmapheresis (DFPP) in kidney transplant recipients

Study Groups

RCA-DFPP (n= 62 sessions)

Hep-DFPP (n= 50 sessions)

Inclusion Criteria

Kidney transplant recipients (≥21 years old) who received DFPP treatment for either pretransplant desensitization or as treatment for antibody-mediated rejection (AMR)

Exclusion Criteria

Not explicitly stated

Methods

DFPP was performed using an Infomed HF440 machine with Granopen 060 plasma filter and Medopen 10 plasma fractionator. For RCA-DFPP, blood flow rate was set to 120-150 mL/min, with anticoagulant citrate dextrose solution A (ACD-A) administered at Y-site connector at a start citrate dose of 2.5 mmol/L. For Hep-DFPP, blood flow rate was set at 120-250 mL/min, with initial heparin bolus of 2000 IU followed by 1000 IU/h or reduced dose (500 to 1000 IU then 250 to 500 IU/hour). Selection of anticoagulation was per clinician's discretion, although RCA-DFPP was preferred for patients with increased risk of bleeding. 

Duration

February 2018 to May 2021

Outcome Measures

Primary: Premature circuit clotting

Secondary: Adverse events 

Baseline Characteristics  

RCA-DFPP (n= 62)

Hep-DFPP (n= 50)

p-value

Age, year

42  52  0.97

Female

45 (72.6%) 33 (66%) 0.45

Primary kidney disease

Glomerulonephritis

Diabetic nephropathy

Hypertensive nephropathy

Polycystic kidney disease

 

47 (75.8%)

3 (4.8%)

4 (6.5%)

8 (12.9%)

 

39 (78%)

3 (6%)

8 (16%)

0

0.57

Duration of DFPP, minutes

156 120 <0.01

Blood flow rate, mL/min (range)

120 (120-150) 150 (150-200) <0.01

Citrate flow rate, mL/min

2.6 NA -
Results   RCA-DFPP (n= 62) Hep-DFPP (n= 50)

p-value

Premature circuit clotting

10 (16.1%) 3 (6%) 0.10

Catheter malfunction with premature circuit clotting

4 (6.5%) 0 0.01

Bleeding episode

3 (4.8%) 0 0.25

Post-DFPP electrolyte abnormalities

Hypocalcemia

Hypomagnesemia

Metabolic acidosis

Metabolic alkalosis:

Hypernatremia

 

3 (4.8%)

30 (48.4%)

20 (32.3%)

2 (3.2%)

1 (1.6%)

 

22 (44%)

11 (22%)

26 (52%)

1 (2%)

0

 

< 0.01

0.03

0.04

0.68

0.36 

Adverse Events

See above; hypotension, bleeding episodes, electrolyte imbalances (hypocalcemia, hypomagnesemia, metabolic acidosis)

Study Author Conclusions

RCA-DFPP may be comparable to Hep-DFPP for maintaining circuit patency in kidney transplant patients. Functional vascular access is crucial to avoid premature clotting. Monitoring for electrolyte imbalances and coagulopathy is recommended.

Critique

The study provides valuable insights into anticoagulation strategies for DFPP, highlighting the importance of vascular access and monitoring. However, the retrospective design and small sample size may limit the generalizability of the findings. The choice of anticoagulation was at the discretion of the treating physicians, potentially introducing bias.

 

References:

Teh SP, Ho QY, Kee YST, et al. Regional citrate anticoagulation vs systemic heparin anticoagulation for double-filtration plasmapheresis. J Clin Apher. 2023;38(1):16-23. doi:10.1002/jca.22019

 

Membrane therapeutic plasma exchange with and without heparin anticoagulation

Design

Retrospective study

N= 1,034 patients (9,611 procedures)

Objective

To evaluate the safety of different types of heparin anticoagulation in therapeutic plasma exchange (TPE)

Study Groups

Group 1 (UH): n= 7,733 procedures

Group 2 (LMWH): n= 575 procedures

Group 3 (No anticoagulation): n= 1,193 procedures

Inclusion Criteria

Patients undergoing TPE at the University Hospital Center Zagreb 

Exclusion Criteria

Citrate use or anticoagulant unknown

Methods

Patient data were retrospectively analyzed from a Croatian hospital. TPE procedures were grouped by anticoagulation type (UH, LMWH, no anticoagulation). Standard protocols were used for dosing. For UH, dose was determined by 50 IU x body weight (kg) + 1000 IU/h. For LMWH, patients were given nadroparin 65 IU/kg, enoxaparin 100 IU/kg, dalteparin 65 IU/kg, or reviparin 50 IU/kg. 

Duration

February 1982 to December 31, 2014

Follow-up: 48 hours post TPE

Outcome Measures

Primary: Incidence of complications during TPE

Secondary: Incidence of blood clotting in the extracorporeal circuit, bleeding complications

Baseline Characteristics  

UH (n= 852)

LMWH (n= 110) Saline (n= 178)
Age, years

37 ± 19

36 ± 22 34 ± 22
Male

41.0%

47.3% 40.4%
Results  

Group 1 (UH)

Group 2 (LMWH) Group 3 (Saline) p-value

Total complications

736 (9.5%) 122 (21.2%) 194 (16.3%) <0.001

Clotting of blood in the extracorporeal circuit

183 (2.4%) 69 (12.0%) 75 (6.3%) <0.001

Inadequate vascular access

152 (2.0%) 21 (3.7%) 23 (1.9%) 0.02

Clotting of blood in extracorporeal circuit, only TPE with adequate vascular access

171 (2.3%) 63 (11.4%) 70 (6.0%) <0.001

Device complications

68 (0.9%) 8 (1.4%) 12 (1.0%) 0.44
Adverse Events

Complications in 11.1% of procedures; severe adverse reactions in 0.6%; blood clotting in 3.4% of procedures; bleeding complications in 0.1% of procedures

Study Author Conclusions

TPE can be conducted safely with UH and, when necessary, without anticoagulation. LMWH was associated with more complications compared to UH and no anticoagulation.

Critique

Strengths: Large sample size, long observation period. Limitations: Single-center, retrospective design, potential referral bias, changes in equipment and protocols over time, lack of randomization.

 

References:

Brunetta Gavranić B, Bašić-Jukić N, Premužić V, Kes P. Membrane therapeutic plasma exchange with and without heparin anticoagulation. J Clin Apher. 2017;32(6):479-485. doi:10.1002/jca.21544

 

Acid citrate dextrose formula A versus unfractionated heparin for anticoagulation of salvaged red blood cells in cardiac surgery

Design

Retrospective, single-center, observational study

N= 8

Objective

To compare patients who underwent red cell salvage and autotransfusion in the course of cardiac surgery, with either acid citrate dextrose formula A (ACD-A) or unfractionated heparin (UFH) used as the anticoagulant for salvaged red blood cells (RBC)

Study Groups

ACD-A (n= 4)

UFH (n= 4)

Inclusion Criteria

Patients who underwent cardiac surgery; underwent red cell salvage and autotransfusion

Exclusion Criteria

None stated

Methods

This was a retrospective study from a single center in Switzerland. All patients underwent general anesthesia and cardiac surgery on cardiopulmonary bypass using a minimally invasive extracorporeal circuit. During weaning, patients received vasoactive and inotropic drug support. Re‐transfusion of processed autologous blood was started only after separation from the extracorporeal circuit. Reinfusion was gravity-driven via a free‐flowing transfusion line. RBC salvage, processing, and autotransfusion were performed using either the autoLogT IQ autotransfusion system (Medtronic) or the CATSmartdevice (Fresenius Kabi).

To prevent blood clotting during the collection period, either ACD-A or UFH was used as an anticoagulant, per guideline recommendations. The ACD-A solutions (from Bichsel AG) contained 22.0 g sodium citrate, 24.5 g glucose monohydrate, and 8 g of citric acid dissolved in 1000 mL of distilled water. Heparan anticoagulation used 25,000 IU of UFH added to a 1000 mL bag of normal saline.

Duration

January to February 2020

Outcome Measures

Mean arterial pressure (MAP), central venous pressure (CVP), and heart rate (HR)

Baseline Characteristics

Of the eight patients, seven underwent coronary artery bypass grafting, and one underwent pulmonary valve replacement.

Mean re‐transfused volume was 247 ± 81 mL.

Results

All four patients in the ACD-A group experienced a decrease of MAP immediately with auto-transfusion (median change, –19 mmHg [range, –29 to –1 mmHg]).

For emergent symptomatic treatment, re‐transfusion had to be interrupted in all ACDA patients, and three of them required vasopressor support.

Patients with UFH‐processed RBC transfusion experienced only clinically insignificant MAP responses (change, +0.75 mmHg [range, −1.5 to +5 mmHg]).

CVP changed inconsistently during autologous reinfusion of ACDA‐salvaged RBC (change, −1 mmHg [range, −15 to +5 mmHg]). In the UFH group, however, CVP remained stable (change, ±0 mmHg [range, −1 to +0.5]).

Heart rate remained stable due to atrial pacing in six of the eight patients. In patient #1 (ACD-A group), heart rate increased by 20 bpm during autologous RBC transfusion. 

In the ACD-A group, five washing cycles were analyzed in four patients. Their RBC concentrates showed extremely low pH (group median, <6.30 [range, <6.30 to 6.49]), a very low ionized calcium concentration of 0.21 mmol/L, and high glucose levels of 23 mmol/L (414 mg/dL). Additionally, lactate levels in this group exceeded serum reference values at 47.75 mg/dL (5.3 mmol/L).

In the UFH group, the median pH was 7.82 [range, 7.79 to 7.93]; ionized calcium, 0.39 mmol/L; glucose, 20.7 mg/dL (1.15 mmol/L)]; and lactate, 38.29 mg/dL (4.25 mmol/L).

Chloride levels (overall median, 135 mmol/L [range, 120 to 146]) were increased in both groups due to the composition of the washing solution.

Adverse Events

See results

Study Author Conclusions

Acid citrate dextrose formula A (ACD-A) anticoagulant for autologous red blood cell salvage has the potential to cause major adverse hemodynamic events during free‐flowing re‐transfusion of autologous red blood cell concentrate. Acute ionized hypocalcemia and acidemia may ensue from residual citrate in the supernatant of red blood cell concentrate reconstituted in unbuffered saline.

InpharmD Researcher Critique

This data is based on consecutive patients, with no explanation of why ACD-A or UFH were used, which opens the possibility of selection bias. The sample size was small and limited to a single center in Switzerland. Limited clinical data or results were presented.



References:

Erdoes G, von Stegmann Und Stein C, Eberle B, Gerber D. Acid citrate dextrose formula A versus unfractionated heparin for anticoagulation of salvaged red blood cells in cardiac surgery. J Card Surg. 2022;37(12):5608-5612. doi:10.1111/jocs.17173

 

Acid-citrate-dextrose Formula A versus heparin as primary catheter lock solutions for therapeutic apheresis

Design

Retrospective chart review

N= 3020 procedures

Objective

To compare efficacy of unfractionated heparin (UFH) and acid-citrate-dextrose Formula A (ACD-A) for therapeutic apheresis (TA)

Study Groups

UFH (n= 295 total courses, 1880 procedures)

ACD-A (n= 132 total courses, 1140 procedures)

Inclusion Criteria

Underwent TA with primary lock solution

Exclusion Criteria

Therapeutic courses excluded if < 3 procedures performed, access other than central venous catheter used, more than one lock solution used throughout course of procedure

Methods

Patient data were compiled via retrospective chart review. All catheters were either temporary (rigid) or tunneled, indwelling hemodialysis type. Specific make/model were not recorded. 

Duration

Underwent procedure between July 2009 to March 2012

Outcome Measures

Patent, partial/total occlusion, tissue plasminogen activator (TPA) used, catheter-related blood stream infection (CRBSI), catheter removal

Baseline Characteristics

 

UFH

ACD-A

   

Total number of courses

Short-term

Long-term

Catheter days > 30

295

225

70

20

132

93

39

12

   

Mean number of procedures

6.4 8.6    

Mean catheter days

11.7  19.2    

Mean dwell time, days

1.8 2.2     

Results

Endpoint

Heparin (n= 294 patients, 1843 procedures)

ACD-A (n= 131 patients; 1134 procedures)

Difference

p-Value

Patent

1617 (87.7%) 934 (82.4%) 5.3 <0.001

Occlusion

Partial

Total

 

214 (11.6%)

 

174 (15.3%)

8 (0.7%)

 

3.7

0.7

 

0.003

<0.001

TPA used

2 (0.1%) 23 (2.0%) 1.9 <0.001

CRBSI

0 1 (0.1%) 0.1 0.381

Catheter removal

1 (0.1%) 1 (0.1%) 0 1

Overall catheter-related outcomes differed by not more than 5.3% for the primary analysis and when stratified by short-term ( 10 days) duration or short dwell times (<3 days). When stratified by long-term duration (>10 days) and long dwell times (>3 days), differences increased to not more than 10.4 and 22.4%, respectively.

Adverse Events

N/A

Study Author Conclusions

For short-term courses and short dwell times, UFH and ACD-A appear equally effective; UFH appears superior to ACD-A in the setting of long-term courses and long dwell times. Major catheter-related complications were rare and occurred with similar frequency in both groups. For most indications, ACD-A appears to be a reasonable alternative to heparin; however, an adequately powered, randomized trial would be required to definitively address this issue.

InpharmD Researcher Critique

The present study is limited due to its retrospective design and heterogeneity between groups. Some outcomes, such as catheter patency, were subjective and may have introduced bias or inconsistency, particularly due to the retrospective study design. 



References:

Osby M, Barton P, Lam CN, Tran MH. Acid-citrate-dextrose Formula A versus heparin as primary catheter lock solutions for therapeutic apheresis. Transfusion. 2014;54(3):735-743. doi:10.1111/trf.12310