[Effect of different connection schemes of continuous renal replacement therapy and extracorporeal membrane oxygenation on arterial and venous pressure: an in vitro and in vivo study].

Objective: To investigate the effects of different connection schemes of continuous renal replacement therapy (CRRT) and extracorporeal membrane oxygenation (ECMO) on arterial pressure (PA), venous pressure (PV), and transmembrane pressure (TMP), and to provide a theoretical basis for choosing a suitable connection scheme.

Methods: (1) In vitro study: the different connection schemes of CRRT and ECMO were simulated and divided into 6 schemes according to the connection between CRRT and ECMO circuits at different positions. Scheme A: connected to the front and back points of the oxygenator; scheme B: connected to the points behind and in front of the oxygenator; scheme C: connected to the points in front of the oxygenator and in front of the centrifugal pump; scheme D: connected to the points behind the oxygenator and in front of the centrifugal pump; scheme E: connected to the points in front of the oxygenator and the return catheter; scheme F: connected to the points after the oxygenator and the return catheter. Each set of ECMO circuits was measured 5 times under each connection scheme and different flow rates (2, 3, 4, 5, 5.5 L/min). Six ECMO circuits for a total of 30 measurements, and the PA, PV, and TMP of the 6 schemes were compared. (2) In vivo study: the patients who were treated with ECMO combined with CRRT in the department of critical care medicine of the First Affiliated Hospital of Wannan Medical College from August 2017 to August 2021 changed the connection scheme due to high PA or PV (from scheme A or B to scheme E or F) were retrospectively analyzed. The changes of PA and PV before and after changing the scheme were compared.

Results: (1) In vitro study results: there was no significant difference in PA between schemes A and B, C and D, E and F under different ECMO blood flow (2-5.5 L/min). The PA of schemes C and D was the lowest, followed by schemes E and F. PV of scheme B was higher than that of scheme A under different ECMO blood flow (2-5.5 L/min). There was no significant difference in PV between schemes C and D, E and F under high ECMO blood flow (3-5.5 L/min), and the absolute value of PV was lowest in schemes E and F. Compared with schemes A and B [partial PA > 300 mmHg (1 mmHg ≈ 0.133 kPa) at high flow rate], C and D (partial PV > 350 mmHg at high flow rate), schemes E and F were more reasonable connection schemes. TMP was negative in schemes C and D at ECMO blood flow of 5 L/min and 5.5 L/min (mmHg; 5 L/min: scheme C was -29.14±11.42, scheme D was -42.45±15.70; 5.5 L/min: scheme C was -35.75±13.21, scheme D was -41.58±15.42), which indicated the presence of dialysate reverse filtration. Most of the differences in TMP among schemes A, B, E, and F under different ECMO blood flow (2-5.5 L/min) were statistically significant, and the absolute value of mean fluctuation was 9.89-49.55 mmHg, all within the normal range. (2) In vivo study results: a total of 10 patients who changed the connection scheme (from scheme A or B to E or F) due to high PA or PV were enrolled, including 8 males and 2 females; 7 cases of venous-arterial ECMO (VA-ECMO) and 3 cases of venous-venous ECMO (VV-ECMO), all used continuous veno-venous hemodiafiltration (CVVHDF) mode. After changing the scheme, both PA and PV decreased significantly as compared with those before changing [PA (mmHg): 244.00±22.58 vs. 257.20±21.92, PV (mmHg): 257.20±18.43 vs. 326.40±15.41, both P < 0.01], and PV decreased more significantly than PA [difference (mmHg): 69.20±6.55 vs. 13.20±5.45, P < 0.01].

Conclusion: For patients treated with ECMO in combination with CRRT, the scheme of connecting the access line of CRRT to the pre-oxygenator or post-oxygenator and connecting the return line to the point of the return catheter can significantly reduce PA and PV and maintains normal CRRT operation even running high-flow ECMO.