Pharmacotherapy Management in Patients with Extracorporeal Membrane Oxygenation

Ayesha Ather, PharmD, BCPS, BCCPProgram Director, PGY2 Cardiology Pharmacy Residency

Assistant Adjunct Professor, College of PharmacyUniversity of Kentucky

Faculty Disclosure

• I have no conflicts of interest to disclose.

• Gap = Lack of treatment guidelines and published research often leave providers with no clear way to optimally treat patients

• Need = Our learners need strategies to manage patients on extracorporeal membrane oxygenation (ECMO)

Educational Need/Practice Gap

Upon completion of this educational activity, you will be able to:

1. Identify alterations in pharmacokinetics (PK) associated with ECMO

2. Review dose adjustments and monitoring of analgesics, sedatives, and antimicrobials in critically ill patients on ECMO

3. Evaluate the anticoagulation management and monitoring practices in patients on ECMO

Objectives

• What is the desired change/result in practice resulting from this educational intervention?

• As a result of the information/tools provided in this activity, learners should be better able to utilize appropriate pharmacologic therapies to manage patients on ECMO

Expected Outcome

ECLS Registry Report: International Summary 2019

Pharmacokinetic Alterations

Drug Factors Disease Factors

Extracorporeal Factors

Critical Illness

Augmented Cardiac Output

Leaky Capillaries/Volume

Resuscitation

Altered Protein Binding

End-organ Dysfunction

Increased Clearance Increased Volume of Distribution

Decreased Plasma

Concentrations

Decreased Clearance

Increased Plasma

Concentrations

Dzierba et al. Crit Care. 2017:21;21(1):66

Extracorporeal Membrane Oxygenation

Augmented Cardiac Output Hemodilution Drug Sequestration End-organ

Dysfunction

Increased Clearance Increased Volume of Distribution

Decreased Plasma

Concentrations

Decreased Clearance

Increased Plasma

Concentrations

Dzierba et al. Crit Care. 2017:21;21(1):66

• ECMO Circuit• Tubing type• Oxygenator membrane• Priming solution• Age of the circuit

Extracorporeal Membrane Oxygenation

A: Tubing/PumpB: OxygenatorC: Priming solution

A

BC

Preston et al. J Extra Corpor Technol 2010 S;42(3):199-202Shekar et al. J Crit Care 2012; 27(6): 741.e9-18

Wildschut et al. Intensive Care Med 2010; 36(12): 2109-2116

Drug Factors - Analgesics and Sedatives

Dzierba AL et al. Pharmacotherapy 2019 Mar;39(3):355-368

Lipophilicity (log p value) and protein-binding properties of common opioids and sedatives

Analgesics and Sedatives

0

20

40

60

80

100

120

Morphine Midazolam Fentanyl Propofol

Perc

enta

ge

Simulated Adult ECMO Circuit

0 Minutes1440 Minutes

Shekar et al. Crit Care. 2012;16(5):R194Lemaitre et al. Critical Care. 2015;19:40

How to Manage Pain and Sedation

Routine assessment of pain, agitation and delirium

Pain should be treated before sedation

Target light sedation (vs deep sedation)

Propofol or dexmedetomidine preferred over benzodiazepines for sedation

Performing rehabilitation or mobilization

Key Guideline Concepts

Barr J, et al. Crit Care Med 2013;41(1):263-306Devlin JW, et al. Crit Care Med 2018;46(9):e825-e873

48-hrs post VV ECMO

initiation(n=45)

Deeply sedated, n (%) 43 (96)Continuous infusion sedative, n (%) 43 (96)Continuous infusion opioid, n (%) 44 (98)Daily propofol dose in mg, median (IQR)

3,380 (1,105–4,110)

Daily midazolam equivalents dose in mg, median (IQR)

202 (103–247)

Daily fentanyl equivalents dose in mcg, median (IQR)

4,800 (3,000–5,820)

Application of Guidelines

24-hrs before VV ECMO

discontinuation(n=35) 8 (23)

16 (46)24 (69)1,760

(960–2,960)32

(14–81)1,625

(610–3,345)

48-hrs post VV ECMO

discontinuation (n=30)1 (3)

8 (27)12 (40)

660 (540–2,220)

32 (10–122)

720 (150–1,660)

DeBacker J, et al. ASAIO J 2018;64:544-551

IQR: interquartile range

Initial Preferred Opioid n=221 (%)

Fentanyl 171 (77)

Hydromorphone 36 (16)

Morphine 10 (5)

Current Practice and PerceptionsSecond Preferred

Opioid n=221 (%)

Hydromorphone 106 (48)

Morphine 48 (21)

Fentanyl 38 (17)

Initial Preferred Sedative for Deep

Sedationn=221 (%)

Propofol 155 (70)

Benzodiazepines 54 (24)

Dexmedetomidine 9 (4)

Ketamine 2 (1)

Initial Preferred Sedative for Light

Sedationn=221 (%)

Dexmedetomidine 100 (45)

Propofol 85 (39)

Benzodiazepine infusion 16 (7)

Benzodiazepine prn 9 (4)

Second Preferred Sedative for Deep

Sedationn=221 (%)

Benzodiazepines 90 (41)

Dexmedetomidine 52 (23)

Propofol 41 (19)

Ketamine 27 (12)

Second Preferred Sedative for Light

Sedationn=221 (%)

Dexmedetomidine 81 (37)

Benzodiazepine prn 49 (22)

Propofol 45 (20)

Ketamine 12 (5)

Perception that opioid dosing is higher with VV-ECMO = 121 (55)Perception that sedation dosing is higher with VV-ECMO = 131 (59)

Dzierba et al. J Crit Care. 2019 Oct;53:98-106

• Drug dosing recommendations are unlikely to be evidence-based

• Use published pharmacokinetic data in critically ill patients to make dosage adjustments

• Set daily sedation goals and consider daily interruption of sedatives

• Lipophilicity and protein binding appear to be important factors affecting pharmacokinetics

Analgesia and Sedation Considerations

• Therapeutic failure

• Potential emergence of resistant microorganisms

• Toxicity

Antimicrobial Dosing Considerations

HA et al. Pharmacotherapy. 2017;37(2):221-235

Drug Factors - Antimicrobials

Lipophilicity (log p value) and protein-binding properties of common antimicrobials

Dzierba AL et al. Pharmacotherapy 2019 Mar;39(3):355-368

VancomycinPatients Endpoints

11 ECMO11 Controls No difference in

clearance or volume of distribution

20 ECMO60 Controls11 ECMO11 Controls

Antimicrobials and ECMO

AmikacinPatients Endpoints

50 ECMO50 Controls

No difference in Cmax and Cmin

Cmax <60 mg/mL = 26% (ECMO) vs. 34% (Controls)7 ECMO50 Controls

Patients receiving ECMO had a higher Vd and lower Cl

106 ECMO Cmax <60 mg/mL = 39%Cmax = peak serum concentrations; Cmin = trough serum concentrations; Vd = volume of distribution; Cl = clearance

Gélisse et al. Intensive Care Med. 2016;42(5):946-948Ruiz-Ramos et al. ASAIO J. 2018;64(5):686-688

Touchard et al. Crit Care. 2018;22(1):199Park SJ et al. PLoS One. 2015;10(11):e0141016

Wu CC et al. J Formos Med Assoc. 2016;115(7):560-570Donadello K et al. Crit Care. 2014;18(6):632

• Case control cohort: Total of 41 therapeutic drug monitoring (TDM) results

β-Lactam Pharmacokinetics in ECMO

Meropenem(n=27)

Piperacillin/Tazobactam(n=14)

ECMO Control ECMO ControlVolume of Distribution (L/kg) 0.46 (0.26–0.92) 0.60 (0.42–0.90) 0.33 (0.26–0.46) 0.31 (0.21–0.41)

Elimination half life (h) 3.0 (2.1–4.8) 2.9 (2.4–3.7) 2.0 (1.1–4.2) 1.6 (1.0–4.7)

Total drug clearance (mL/min) 125 (63–198) 144 (97–218) 156 (91–213) 134 (47–179)

Donadello et al. Int J Antimicrob Agents. 2015;45(3):278-82

β-Lactam Pharmacokinetics in ECMO

Donadello et al. Int J Antimicrob Agents. 2015;45(3):278-82

Drug Protein Binding Log p Volume of

DistributionExpected

EffectDose

Adjustment

Ceftriaxone 85-90% -0.01 5.78–13.5 L Moderate sequestration Not required

Vancomycin 50% -4.4 28–70 L Minimal sequestration Not required

Levofloxacin 24–38% 0.65 88.9 LMinimal to moderate

sequestrationNot required

Gentamicin/Tobramycin/Amikacin

< 30% < 0.0 14–21 L Minimal sequestration Not required

Voriconazole 58% 2.56 322 L Moderate to high sequestration Yes

Dose Adjustments for Select Agents

HA et al. Pharmacotherapy. 2017;37(2):221-235

• PK data in adult patients on ECMO are sparse

• Consider loading dose for drugs with moderate to high

sequestration

• Dose guided by therapeutic drug monitoring when applicable

• Monitor for signs of infections

Antimicrobial Dosing Considerations

Bleeding and Thrombosis Complications

• Meta-analysis: 12 studies (1763) patients

• Any bleeding (33%)

• Hemolysis (18%)

• Venous thrombosis (10%)

• Gastrointestinal bleeding (7%)

• Disseminated intravascular coagulation (5%)

Zangrillo et al. Crit Care Resusc. 2013;15(3):172-178

The Clinical Challenge

ThrombosisBleedingWhich

Anticoagulant?

Zangrillo et al. Crit Care Resusc. 2013;15(3):172-178

Guidelines

“These guidelines describe useful and safe practice, but these are not necessarily consensus recommendations. These guidelines are not

intended as a standard of care, and are revised at regular intervals as new information, devices, medications, and techniques become available.”

• Heparin bolus (50-100 units/kg) at time of cannulation, continuous infusion during ECLS

• Direct thrombin inhibitors• Monitor ACT, aPTT, or anti-Xa

GoalsACT 180-200 secMedian antithrombin 70%Anti-Xa 0.3-0.7 IU/mL

Transfusion TriggersPlatelets <100kFibrinogen <145mg/dL

Monitoring FrequencyAPTT q6-8hCBC q6-8hFibrinogen >12hFree hemoglobin >12hAntithrombin q13-24hAnti-Xa q13-24h

Practice Survey of 121 ECMO Centers

Bembea et al. Pediatr Crit Care Med 2013;14(2): e77

Predictable dose response Stable dosing Quick onset Reliable monitoring ReversibilityWithout adverse drug reactions Effectively prevent thrombosis Minimal bleeding risk

Ideal Parenteral Anticoagulant

Pharmacokinetics (PK)Pharmacodynamics (PD)

Usability

Outcomes

Garcia et al. Chest. 2012;141(2 Suppl):e24S-e43S

Bivalirudin

Proteolytic Metabolism +

Renal EliminationDirect Thrombin

Binding

PK and PD

Predictable Onset of Action & EliminationPredictable Dose Adjustments

aPTT Stability

Heparin

Plasma Protein Binding

ATIII Dependence

Unpredictable Onset of Action & EliminationNonlinear Dose Adjustments

aPTT Instability & Heparin Resistance

Thomas et al. Semin Thromb Hemost. 2018;44(1):20-29Coughlin et al. ASAIO J. 2015;61(6):652-655

Heparin

Monitoring ACT, aPTT, anti-Xa, TEG/ROTEM

Reversal ProtamineHalf life 90 min

ADR HIT, thrombocytopenia

Bivalirudin

Monitoring ACT, aPTT, TT, TEG/ROTRM

Reversal No antidoteHalf life 25 min

ADR Minimal

Heparin vs Bivalirudin

Thomas et al. Semin Thromb Hemost. 2018;44(1):20-29Coughlin et al. ASAIO J. 2015;61(6):652-655

ACT: activated clotting time; aPTT: activated partial thromboplastin time; TEG or ROTEM, thromboelastography or rotational thromboelastometry; TT: thrombin time; ADR: adverse drug reaction

Study Design Subjects Bivalirudin Outcomes

Ranucci, 2011 Retrospective21 (11 adults) PostcardiotomyVA-ECMO

n=13ACT 160-180 secaPTT 50-80 sec

Blood product use: FFP, plateletsBlood loss

Pieri, 2013 Retrospective 20 adultsVA-ECMO 10

n=10aPTT 45-60 sec

Major/Minor bleeding = NDThrombosis = ND

Berei, 2017 Retrospective 72 adultsVA-ECMO 66

n=44aPTT 45-65 sec or 60-80 sec

Major/Minor bleeding = NDThrombosis = ND

Heparin vs Bivalirudin

ND: No difference

Bivalirudin

Heparin

Heparin

• Most data in pediatric population

• Center specific protocols

• Heparin is the drug of choice???

• Variable monitoring strategies

• UK primarily uses bivalirudin and monitors aPTT

Anticoagulation Considerations

Pharmacotherapy Management in Patients with Extracorporeal Membrane Oxygenation

Ayesha Ather, PharmD, BCPS, BCCPProgram Director, PGY2 Cardiology Pharmacy Residency

Assistant Adjunct Professor, College of PharmacyUniversity of Kentucky