Prevenzione della disfunzione endoteliale in corso di sepsi: ruolo della CPFA (Coupled Plasma Filtration Adsorption)


Sepsis has been recognized as a common cause of acute kidney injury (AKI), and the combination of these two stress conditions has been associated with very high morbidity and mortality (Zarjou A-2011) [1] (full text).

The pathogenesis of sepsis-induced AKI involved endothelial dysfunction, infiltration of inflammatory cells in the renal parenchyma, intra-glomerular thrombosis, and apoptosis/necrosis of tubular cells (Zarjou A-2011 [1] (full text)).

The term “endothelial dysfunction” (Wu L-2007) [2] (full text) referred to functional and phenotypic changes that lead to a cellular switch from a quiescent to a dysfunctional activated state (Guerrot-2012 [3] (full text).

The use of sorbents, typically resins, in extracorporeal therapy in particular CPFA, has revealed an improvement in the procedures of blood detoxification and the survival of septic subjects (Mao HJ-2009) [4].

The aim of our study was to evaluate the CPFA efficacy in preventing the effects of sepsis mediators on endothelial dysfunction.


After 3 h from LPS infusion, 8 pigs were treated with CPFA for 6 h; 8 control pigs receive no treatment. Renal biopsies were performed before (T0) and 9 hours (T9) after LPS infusion. Human ECs were co-incubated for 12 h with different swine sera and analyzed by flow cytometry.


By flow cytometry analysis, ECs stimulated with sera of healthy pigs (CTR sera) showed high levels of endothelial markers CD31 (93.57 ± 6.11 MFI) and VE-cadherin (26.43 ± 2.11 MFI), and low levels of endothelial dysfunction markers N-cadherin (0.43 ± 0.05 MFI) and Vimentin (34.53 ± 5.61 MFI)(Figure 1, A-B).

ECs cultured with septic sera (LPS sera) showed reduced expression of specific ECs markers (CD31 49±9.5 vs CTR sera p = 0.04) and VE-cadherin (11.1±2.3 vs CTR sera p = 0.01), with increased expression of markers of ECs dysfunction such as N-cadherin (22.3 ± 6.6 vs CTR sera p = 0.02) and Vimentin (59.23 ± 13 vs CTR sera p = 0.02). Surprisingly sera of CPFA septic pigs (LPS+CPFA sera) preserved ECs phenotype (CD31:94.97 ± 3.4 p = 0.04; VE-cadherin: 24.07 ± 4.9 p = 0.03 vs LPS sera) without up-regulation of N-cadherin (0.97 ± 0.14 p = 0.02 vs LPS sera) and Vimentin (28.17 ± 6.77 p = 0.02, vs. LPS sera) (Figure 1 A-B).

The analysis of septic renal biopsies (T9 LPS) also demonstrated in vivo the dysfunction of ECs, which acquired the expression of myofibroblast marker a-SMA (B) compared to T0 (A) (CD31+/α-SMA18,09 ± 1.58 vs T0:1,33 ± 0.57, p 0.005) (Figure 2, A-C). Zoomed image showed an ECs (Figure 2, F) co-expressing CD31 (D) and α-SMA (E) in septic tissue. Otherwise, early treatment by CPFA (T9 LPS CPFA) was able to retain ECs morphology and to reduce α-SMA+ ECs (CD31+/α-SMA+ 5.58 ± 0.52 vs. septic, p=0.01) (Figure 3 B, D, E).


Our data indicated that CPFA treatment might be pivotal to counteract ECs dysfunction both in vitro and in vivo, by removing cytokines involved in ECs phenotypic and functional changes.