ECs exposed to pdRBCs and sRBCs demonstrate increased endothelial inflammation within the smallest channels and at bifurcations in microfluidic devices mimicking small venules. (A) Macroscopic view of a microfluidic device designed to mimic a small-medium venule, in which the smallest channels are ∼30 μm in diameter. (B) Brightfield microscopy image at 10× original magnification of the initial branchpoint of our small-medium venule-mimicking microfluidic device endothelialized to confluence with human umbilical vein ECs. (C) Endothelial VCAM-1 (green) and E-selectin (red) expression in the first branchpoint of the microfluidic device after 4-hour perfusion of healthy control RBCs vs RBC suspensions containing varying amounts (5%, 10%, and 100%) of sRBCs (left) and nystatin-treated, less-deformable RBCs (right). All RBC suspensions were diluted in media to a hematocrit of 25% and perfused at a constant venular shear rate. (D) Graphical representation of mean normalized fluorescent intensity for VCAM-1 and E-selectin, both markers of inflammation, in ECs exposed to hRBCs (red) and RBC suspensions with varying amounts of sRBCs (left) and nystatin-treated, less-deformable RBCs (right) at the initial branchpoint shown. A total of 5 separate experiments were analyzed. VCAM-1 and E-selectin expression increased as the number of sickled RBCs increased. In the experiments using nystatin-treated RBCs, VCAM-1 and E-selectin had a statistically significant increase in expression as the number of less-deformable cells increased (VCAM-1 2.29 ± 0.46 standard error of the mean [SEM; P < .05]; E-selectin 1.7 ± 0.16 SEM [P < .05]) normalized fluorescent intensity in ECs exposed to 100% nystatin-treated RBCs), indicating that increased RBC-EC interactions occur at vessel branchpoints in the presence of less-deformable RBCs, which also exert increased compressive mechanical forces against the endothelium in the smallest vessels. Statistical analyses using a Mann-Whitney U test; ∗P ≤ .05. Error bars represent SEM.