Effects of continuous renal replacement therapy on intestinal mucosal barrier function during extracorporeal membrane oxygenation in a porcine model

Abstract

Backgrounds: Extracorporeal membrane oxygenation (ECMO) has been recommended for treatment of acute, potentially reversible, life-threatening respiratory failure unresponsive to conventional therapy. Intestinal mucosal barrier dysfunction is one of the most critical pathophysiological disorders during ECMO. This study aimed to determine whether combination with CRRT could alleviate damage of intestinal mucosal barrier function during VV ECMO in a porcine model.

Methods: Twenty-four piglets were randomly divided into control(C), sham(S), ECMO(E) and ECMO + CRRT(EC) group. The animals were treated with ECMO or ECMO + CRRT for 24 hours. After the experiments, piglets were sacrificed. Jejunum, ileum and colon were harvested for morphologic examination of mucosal injury and ultrastructural distortion. Histological scoring was assessed according to Chiu's scoring standard. Blood samples were taken from the animals at -1, 2, 6, 12 and 24 h during experiment. Blood, liver, spleen, kidney and mesenteric lymphnode were collected for bacterial culture. Serum concentrations of diamine oxidase (DAO) and intestinal fatty acid binding protein (I-FABP) were tested as markers to assess intestinal epithelial function and permeability. DAO levels were determined by spectrophotometry and I-FABP levels by enzyme linked immunosorbent assay.

Results: Microscopy findings showed that ECMO-induced intestinal microvillus shedding and edema, morphological distortion of tight junction between intestinal mucous epithelium and loose cell-cell junctions were significantly improved with combination of CRRT. No significance was detected on positive rate of serum bacterial culture. The elevated colonies of bacterial culture in liver and mesenteric lymphnode in E group reduced significantly in EC group (p < 0.05). Compared with E group, EC group showed significantly decreased level of serum DAO and I-FABP (p < 0.05).

Conclusions: CRRT can alleviate the intestinal mucosal dysfunction and bacterial translocation during VV ECMO, which may extenuate the ECMO-associated SIRS and raise the clinical effect and safety.

Figure 1

Histology examination of jejunal, ileal and colonic mucosa with H and E staining under optical microscopy (magnification*100). A-D: jejunal mucosa of C, S, E, EC group; E-H: ileal mucosa of C, S, E, EC group; I-L: colonic mucosa of C, S, E, EC group. Normal mucosa was viewed in C group. Mucosal structure distortion, villous collapse and edema, and epithelial shedding were observed in S and E groups. These damages were significantly attenuated in EC group.

Figure 2

Pathological score of jejunal, ileal and colonic mucosa injury. Socres of S and E group were markedly higher than C group (P < 0.01). Score of EC group was significantly lower than S and E group, but higher than C group (P < 0.01). (Compared to control group, ^*P < 0.01, ^★P < 0.05; compared to S group, ^◆P < 0.01; compared to E group, ^#P < 0.01).

Figure 3

Ultrastructural detection of jejunal, ileal and colonic mucosa under transmission electron microscopy (magnification*10,000). A-D: jejunal mucosa of C, S, E, EC group; E-H: ileal mucosa of C, S, E, EC group; I: bacterial invasion into ileal mucosa; J-M: colonic mucosa of C, S, E, EC group. The morphological distortion of tight junction between intestinal mucous epithelium became blurred with loose cell-cell gaps, disappearance of desmosomes and mitochondrial swelling in E group were significantly improved in EC group.

Figure 4

Colonies of bacterial culture in liver, spleen, kidney and mesenteric lymphnode. VV ECMO therapy (E group) induced a significant increase of colonies of bacterial culture in liver and mesenteric lymphnode than C group (P < 0.01), which was attenuated significantly in liver (P < 0.01) and partially in mesenteric lymphnode after combination with CRRT therapy (EC group). (Compared to control group, *P < 0.01; compared to S group, ◆P < 0.05; compared to E group, #P < 0.05).

Figure 5

Serum DAO levels in 4 groups. Serum DAO levels did not change significantly over time (P > 0.05). Compared with notable increase of serum DAO level in E group, it was significantly lower after CRRT for 12 h (P < 0.01). (Compared to control group, *P < 0.05, **P < 0.01; compared to S group, ◆P < 0.05, ◆◆P < 0.01; compared to E group, #P < 0.05, ##P < 0.01).

Figure 6

Serum I-FABP levels in 4 groups. Serum I-FABP levels did not change significantly over time (P > 0.05). Compared with high serum I-FABP level in E group, it showed significant decrease after CRRT for 12 h (P < 0.05). (Compared to control group, *P < 0.05, **P < 0.01; compared to S group, ◆P < 0.05, ◆◆P < 0.01; compared to E group, #P < 0.05, ##P < 0.01).

Figure 7

Filtrate DAO and I-FABP level over time. A: Filtrate DAO and I-FABP level over time; B: Filtrate DAO and I-FABP level over time. The filtrate DAO and I-FABP level remained relatively constant throughout the experiment period. (Compared to -1 h, *P < 0.05).