A framework for drug pharmacokinetics during cardiopulmonary bypass
Conor J O’Hanlon (1); Jacqueline A Hannam (1); Brian Anderson (2,3); Nick Holford (1 )
(1) Department of Pharmacology and Clinical Pharmacology, University of Auckland, New Zealand; (2) Department of Anaesthesia, Starship Children’s Hospital, Auckland, New Zealand; (3) Department of Anaesthesiology, University of Auckland, New Zealand.
Cardiopulmonary bypass (CPB) is an extracorporeal medical procedure that replaces heart and lung function. CPB can alter pharmacokinetic (PK) parameters through haemodilution, haemofiltration, changes in organ function and drug adsorption to the CPB device. Drug adsorption to CPB or extracorporeal membrane oxygenation (ECMO) devices has been investigated for many drugs classes including sedatives, analgesics, antiarrhythmics, anti-infectives and anti-epileptics [1-7]. These studies report the percentage of recovery from the fluid after circulation through an ECMO or CPB device over a specified period. A more mechanistic approach could describe the time course of adsorption based on the time course of drug concentration in the device, association and dissociation of drug to and from the device surface, and maximum binding capacity of the device.
Cefazolin is a renally eliminated beta-lactam antibiotic used for antimicrobial prophylaxis during cardiac surgery. The impact of CPB on cefazolin PK is unclear and is confounded with other PK changes in neonates, infants and children attributed to size, maturation and renal function. Comorbidity of kidney disease and cardiac disease is common and poor renal function may be exacerbated by cardiac surgery and the CPB procedure. Cefazolin PK during cardiac surgery supported by CPB was investigated using an ex vivo study describing drug adsorption to the device, and an in vivo study describing population PK in neonates to children. This combined approach provides a framework where the rate and extent of adsorption to the device was quantified and incorporated into the in vivo PK model.
The aims of this work were to i) develop a PK model to quantify cefazolin adsorption to CPB devices, ii) show that the principles of this PK model can be used as a framework to quantify adsorption of other drugs such as vancomycin, and iii) develop a population PK model for cefazolin in neonates, infants and children undergoing cardiac surgery supported by CPB, that incorporates the adsorption model.
This work comprised of two studies, an ex vivo and an in vivo study.
Ex vivo study: Closed CPB devices (not connected to a patient) were primed and dosed with either cefazolin, vancomycin or a combination, and samples taken at regular intervals over 1 hour of circulation. Twelve experimental runs were conducted using different CPB device sizes (neonate, infant, child and adult), device coatings (Xcoating™, Rheoparin®, PH.I.S.I.O) and priming solutions. Unbound concentrations of cefazolin (n = 225) and vancomycin (n = 100) were quantified using High Performance Liquid Chromatography (HPLC).
In vivo study: Fifty patients aged 3 days – 13 years were recruited from Starship Children’s Hospital, Auckland, New Zealand, who underwent cardiac surgery supported by CPB. Unbound (n = 678) and total (n = 622) cefazolin concentrations were quantified using HPLC. Unbound concentrations were used to describe drug PK and total concentrations used to estimate plasma protein binding parameters.
Data were analysed using NONMEM (ICON Development Solutions, Maryland USA) version 7.5.0 and Wings for NONMEM version 744 (http://wfn.sourceforge.net/).
A one-compartment model with no elimination was chosen to describe cefazolin disposition ex vivo. A kinetic binding model with saturable binding (Bmax), parameterised by Kd (dissociation constant) and T2off (half-time of dissociation), described the adsorption. Device size specific Bmax parameters and coating specific binding parameters (Kd, T2off) were estimated. The maximum binding capacity for the device sizes were neonate 40.0 mg 95% CI [24.3, 67.4], infant 48.6 mg 95% CI [16.0, 80.2], child 77.8 mg 95% CI [54.9, 103] and adult 196 mg 95% CI [191, 199]. The Xcoating™ Kd estimate was 139 mg/L 95% CI [27.0, 283] and the T2off estimate was 98.4 min 95% CI [66.8, 129]. The Rheoparin® and PH.I.S.I.O coatings binding parameter estimates were not distinguishable. The Kd and T2off estimates for both coatings were 0.196 mg/L 95% CI [0.01, 1.99] and 4.94 min 95% CI [0.17, 59.4]. Data were available from four experimental runs using vancomycin. The mechanistic binding model described the data well with no changes required except estimation of vancomycin specific binding parameters. Vancomycin Bmax estimates were 5.2 mg (neonate), 8.0 mg (child) and 13.4 mg (adult), the Kd estimate was 0.01 mg/L and the T2off estimate was 0.7 min.
A two-compartment model was used to describe cefazolin distribution in vivo. A third compartment (CPB compartment) was “switched on” when the CPB machine was connected to the patient. The CPB volume was fixed to the device priming volume and the intercompartmental clearance between the patient and the CPB device was fixed to the flow rate. Drug was eliminated from the patient and from the CPB device via ultrafiltration processes that occur throughout surgery. The kinetic binding parameters from the ex vivo study described cefazolin adsorption to the CPB device. Population parameters were scaled for size using theory based allometry . Normal Fat Mass was used as a metric for size that can account for body composition . Renal function was described using the ratio of estimated glomerular filtration rate (GFR) to normal GFR  using a serum creatinine measurement before and after surgery. Factors for the impact of CPB on each of the structural parameters were estimated. Population parameter estimates were clearance (CL) 20.0 L/h/70 kg 95% CI [16.5, 23.4], intercompartmental CL (Q) 39.4 L/h/70kg 95% CI [23.3, 53.9], volume of distribution (V) of the central compartment (V1) 11.2 L/70kg 95% CI [6.43, 15.5] and V of the peripheral compartment (V2) 24.5 L/70kg 95% CI [18.3, 32.2]. The factors for CPB describing the change in structural parameter at commencement of bypass were CL 0.67 95% CI [0.52, 0.86], V1 1.07 95% CI [0.516, 1.85], Q 0.85 95% CI [0.57, 1.2], V2 1.25 95% CI [0.93, 1.57]. A single saturable binding site best described cefazolin plasma protein binding with a Bmax estimate of 208 mg/L 95% CI [181, 234] and Kd of 28.7 mg/L 95% CI [24.0, 34.0].
Device size and device coating were important covariates to describe the rate and extent of cefazolin adsorption to CPB devices. The full PK model for cefazolin in neonates, infants and children undergoing surgery supported by CPB incorporated the parametric, mechanism-based model describing drug adsorption to the CPB device. CPB reduced CL and increased V above what could be explained by changes in size, maturation, renal function and additional CPB device volume. Adsorption of cefazolin to the CPB device explained part of the variability in drug concentrations. The vancomycin data indicated that the methodology would be appropriate to quantify vancomycin adsorption and therefore be incorporated into a population PK analysis. This framework could be extended to other PK studies involving CPB.
- Kuntz MT, Pereira LM, Matte GS, Connor K, Staffa SJ, DiNardo JA, et al. Sequestration of Midazolam, Fentanyl, and Morphine by an Ex Vivo Cardiopulmonary Bypass Circuit. ASAIO Journal. 2021;67(12):1342-8.
- McDaniel CG, Honeycutt CC, Watt KM. Amiodarone extraction by the extracorporeal membrane oxygenation circuit. Journal of ExtraCorporeal Technology. 2021;53(1):68.
- Shekar K, Roberts JA, Mcdonald CI, Ghassabian S, Anstey C, Wallis SC, et al. Protein-bound drugs are prone to sequestration in the extracorporeal membrane oxygenation circuit: results from an ex vivo study. Critical Care. 2015;19(1):164.
- Wildschut E, Ahsman M, Allegaert K, Mathot R, Tibboel D. Determinants of drug absorption in different ECMO circuits. Intensive care medicine. 2010;36(12):2109-16.
- Mehta NM, Halwick DR, Dodson BL, Thompson JE, Arnold JH. Potential drug sequestration during extracorporeal membrane oxygenation: results from an ex vivo experiment. Intensive care medicine. 2007;33(6):1018-24.
- Harthan AA, Buckley KW, Heger ML, Fortuna RS, Mays K. Medication adsorption into contemporary extracorporeal membrane oxygenator circuits. Journal of Pediatric Pharmacology Therapeutics. 2014;19(4):288-95.
- Kalaria SN, Kishk OA, Gopalakrishnan M, Bagdure DN. Evaluation of an ex-vivo neonatal extracorporeal membrane oxygenation circuit on antiepileptic drug sequestration. Perfusion. 2021:02676591211028183.
- Anderson BJ, Holford NH. Mechanism-based concepts of size and maturity in pharmacokinetics. Annual Review of Pharmacology and Toxicology. 2008;48:303-32.
- Holford NH, Anderson BJ. Allometric size: the scientific theory and extension to normal fat mass. European Journal of Pharmaceutical Sciences. 2017;109:S59-S64.
- O’Hanlon CJ, Holford NH, Sumpter A, Al‐Sallami HS. Consistent Methods for Fat Free Mass, Creatinine Clearance and Glomerular Filtration Rate to describe Renal Function from Neonates to Adults. CPT: Pharmacometrics and Systems Pharmacology. 2023;12(3):401-12.