A . Chevalier et al .: J Extra Corpor Technol 2024 , 56 , 101 – 107 103
Figure 1 . Schematic of ECMO and CRRT ex-vivo circuit configurations . ( A ) ECMO circuit experiment demonstrating layout of reservoir , tubing , pump and oxygenators and locations of injection and sampling sites . ( B ) CRRT circuit experiment demonstrating CVVHDF configuration as well as location of injection and sampling sites . PBP = pre-blood pump .
Control setup
For the ketamine experiments , 50 ml of the ECMO prime solution was placed in a 50 ml polypropylene centrifuge tube ( CELLTREAT , Pepperell , MA ) to serve as a control solution to measure medication stability over the time course of the experiments . The solution was incubated in a water bath at 37 degrees C . For the dexmedetomidine experiments , data from ECMO circuit ex-vivo experiments previously published by our group served as the control . In these experiments , dexmedetomidine concentrations were demonstrated to be stable over time , with 102 % recovery after 24 h [ 15 ].
Medication administration and sample collection
Ketamine and dexmedetomidine were obtained from the University of Utah hospital pharmacy . Medication was dosed to the circuit experiments and the control experiments targeting therapeutic concentrations for ketamine ( 200 ng / ml ) [ 22 ] and dexmedetomidine ( 1.6 ng / ml ) [ 23 ].
For the ECMO circuits , medication was administered via a three-way stopcock in the venous limb ( Figure 1A ). Blood samples were collected from the arterial limb of the circuit 1,5,15 , and30min , and1,2,3,4,5,6 , and8haftermedication administration . Experiments were performed in triplicate .
For the CRRT circuits , medication was administered via a three-way stopcock on the access line downstream from the reservoir . Blood samples were drawn from a stopcock via the return line just upstream from the reservoir and hemofiltrate samples were drawn from a stopcock just upstream from the effluent bag ( Figure 1B ). Paired blood and hemofiltrate samples were collected at 1 , 5 , 15 , and 30 min , and 1 , 2 , 3 , 4 , 5 , and 6 h after medication administration . Due to multiple premature circuit failures for the ketamine experiments , three runs were performed out to 6 h , one circuit ran for 5 h , and one circuit ran for 2 h . Dexmedetomidine experiments were performed in triplicate .
Medication was directly added to the control experiments followed by 5 min of mixing on a rotary mixer prior to sample collection . Samples were subsequently collected at 1 , 5 , 15 , and 30 min , and 1 , 2 , 3 , 4 , 5 , 6 , and 8 h , with gentle inversion prior to sample collection to ensure adequate mixing .
All blood samples were centrifuged at 3000 g at 4 degrees C for 10 min . The plasma was manually pipetted off and immediately frozen in cryovials ( Fisher Scientific , Pittsburgh , PA ) and stored at �80 degrees C until analysis . Hemofiltrate samples were pipetted directly into cryovials ( Fisher Scientific , Pittsburgh , PA ) and stored at �80 degrees C until analysis .
Sample analysis
Plasma and hemofiltrate samples for the ketamine and dexmedetomidine experiments were analyzed at OpAns Laboratory ( Durham , NC ) using high-performance liquid chromatography combined with tandem mass spectrometry ( HLPC – MS / MS ). The range of quantitation for the ketamine and dexmedetomidine assays was 1 ng / ml-10,000 ng / ml and 25 pg / ml-10,000 pg / ml respectively .
Percent recovery calculation
To correct for variations in initial drug concentration due to small differences in circuit volume , drug percent recovery for each sample collection was calculated using the following equation :
Recovery ð% Þ ¼ C t
100 %
C ref
where C t was the concentration at time t , and C ref was the reference concentration of the drug in the circuit . For both the ECMO experiments and the CRRT experiments , medication was not detected in the plasma samples at 1 minute as circuit mixing was not immediate . As a result , the concentration at 5 min was used as the reference concentration for both the circuit experiments and the control experiments .
Hemofiltrate calculations The saturation coefficient for CRRT was calculated using the following equation :
S a ¼ C eff
C p
where S a is the saturation coefficient , C eff is the effluent concentration , and C p is the plasma concentration . The saturation coefficient was calculated at each timepoint from 15 min onwards as this allowed for time for medication equilibration throughout the hemofiltrate .