162 C . C . Honeycutt et al .: J Extra Corpor Technol 2023 , 55 , 159 – 166
Figure 2 . Recovery of meropenem in ECMO circuit configurations and controls over 24 h after administration . Values shown are means ( n = 3 ) with error bars representing standard deviations . ECMO – extracorporeal membrane oxygenation .
( 50 – 100,000 ng / mL ). The LLOQ for meropenem was 50 ng / mL and the accuracy ranged from 94.6 % to 103.5 %.
Due to differences in ECMO and CRRT circuit volumes , meropenem concentrations varied slightly between experiments . Drug recovery was therefore calculated using the following equation :
Recovery ð % Þ¼ C t
100 ;
C i
where C t is the concentration at time = t and C i is the initial concentration . The initial concentration was set at time = 5 min , as there was inadequate mixing in ECMO and CRRT circuits at time = 1 min . Data are reported as mean and standard deviation ( SD ).
In CRRT , drug passage across the hemofilter was calculated from paired hemofiltrate and plasma samples at each time point using the following equation :
SA ¼ C H
;
C P
where SA is the saturation coefficient and C H and C P are concentrations in hemofiltrate and plasma , respectively .
Statistical tests for ECMO comparing all circuit configurations and control were performed at the 24-hour time point using one-way ANOVA . Statistical tests for CRRT comparing circuit and control were performed at the 2-hour time point using a paired t-test . GraphPad Prism Version 9.5.1 and Microsoft Excel were used for statistical analysis and graphing . Further details of statistical analysis and replicates are included in the figure legends . Lines represent the mean and error bars signify the standard deviation ( SD ). A supplementary table of all relevant raw data for ECMO and CRRT is available ( Supplementary Table 1 ).
Results ECMO circuits
All ex vivo ECMO circuit configurations and controls showed a steady decline in meropenem recovery over the course of 24 h ( Figure 2 ). Mean ( standard deviation , SD ) recovery of meropenem at 24 hours was 15.6 % ( 12.9 ) for Complete circuits , 37.9 % ( 8.3 ) for Oxygenator circuits , 47.1 % ( 8.2 ) for Pump circuits and 20.6 % ( 20.6 ) for controls . At 24 h , there were no statistically significant differences in meropenem recovery between Complete , Oxygenator , and Pump circuits , and control ( p = 0.0668 ). Two meropenem concentrations were not included in the analysis due to presumed contamination ( Circuit 6 [ Oxygenator Circuit ] at t = 1 h ; Circuit 9 [ Pump Circuit ] at t = 30min ). CRRT circuits
Meropenem was rapidly cleared by CRRT with mean ( standard deviation , SD ) recoveries of 2.36 % ( 1.44 ) at two hours and 0.13 % ( 0.11 ) at 4 h in circuits compared with a mean ( SD ) recoveries of 93.0 % ( 7.1 ) at two hours and 78.6 % ( 4.5 ) at four hours in controls . The recovery in CRRT circuits was significantly different compared to the control at the two-hour time point ( p < 0.0001 ) ( Figure 3 ). The mean hemofiltration saturation coefficient ( SA ) was approximately one for the duration of the experiments , suggesting free filtering of meropenem from the circuits into the hemofiltrate . The average pH over ECMO and CRRT circuits was 7.37 ( min = 7.12 , max = 7.55 ). pH values outside of the target physiological range ( 7.2 – 7.5 ) were treated as per above .
Discussion
ECMO and CRRT are critically important and lifesaving ECLS modalities . Many patients with severe infections , especially those caused by extended-spectrum b-lactamase ( ESBL ) -producing Enterobacteriaceae , are treated with meropenem [ 1 ]. These patients may also be placed on ECMO and / or CRRT [ 36 ]. As there have been numerous reports of pharmacokinetic alterations and drug extraction in patients with ECLS , it is of interest to determine the extent of meropenem extraction in ECMO and CRRT . It is also of interest to determine the contributions of different circuit components to such drug extraction . In the present study , we demonstrate rapid extraction of meropenem from ex vivo CRRT circuits . We also saw substantial loss of meropenem in ECMO circuits but this loss was