Fig 1.
A) Schematic of the venous and arterial anatomy in the upper arm depicting the brachial artery, the cephalic vein and cephalic arch (CA). B) Venogram showing the straight cephalic vein (Pre-bend; yellow), the bend of the cephalic arch (Bend; red arrow) and straight vein (Post-bend; blue). C, D) Step-by-step processing of C) venogram, and D) IVUS images to reconstruct the 3D genometry of a patient’s CA downstream of AVF placement that are used to construct internal vein contours. Red arrows in the (C) panels indicate the CA. E) Venogram and IVUS data are combined to reconstruct model of the patient’s CA in 3D. F) Reconstructed model shown here is lofted and smoothed to render the 3D model used for flow simulation. G, H) Doppler trace of blood flow in a patients’ CA at 3 and 12 months after BCF placement, respectively.
Fig 2.
Study subject inclusion and discontinuation.
Of 161 subjects enrolled, 5 subjects were suitable and agreed to participate with IVUS measurements at 3 and 12 months. 40 subjects completed the trial, 56 had alternate access placed, 24 had primary failure and never used the access for dialysis, 22 deaths occurred which were unrelated to the study, 14 subjects either received renal transplantation or transitioned to peritoneal dialysis, and 5 subjects were lost to follow-up.
Fig 3.
Three dimensional models of cephalic arches in five ESRD patients obtained 3 months (top) and 12 months (bottom) after their AVF placement.
The models, displayed in the xy-plane, are reconstructed from venogram and IVUS imaging.
Table 1.
Patient characteristics.
Table 2.
Patient measured and calculated parameters.
Fig 4.
A) An idealized model of a CA in ESRD patient. The diameter of the model is 9.5 mm and bend angle, α is 125°. The blue, red and green regions indicate the pre-bend, bend and post-bend regions, respectively. B) Simulated velocity and (C) pressure profiles, respectively in the 3D model under ESRD-specific physiological conditions. D-F) Velocity profiles across the CA model in the (D) pre-bend, (E) bend, and (F) post-bend regions in the idealized model across the tubular cross-section. Different inlet flow velocities range from healthy to ESRD values.
Fig 5.
Panels P93, P96, P98, P104 and P122 show the velocity and pressure profiles in 3D models constructed from IVUS and venogram measurements on the five ESRD patients’ CA at 3 mo. (top), and 12 mo. (middle) and follow-up venogram (bottom) after AVF placement.
The color bar indicates the magnitude of velocity and pressure, respectively following a rainbow scheme, with blue denoting the lowest and red the highest magnitude. Black tubes superimposed on the velocity profiles indicate the velocity field at that point with the tube diameter being proportional to the shear rate. Flow parameters obtained from each patient’s vitals and Doppler data were used to simulate flow.
Fig 6.
Each panel shows the velocity profile in the CA of patient P96 at (A) 3 mo. and (B) 12 mo. under pulsatile flow. In each figure, the first panel shows the inlet velocity as it cycles between peak systolic and diastolic as function of time sampled during pulsatile flow. The color bar in the last panel indicates the velocity magnitude. Black tubes marked on each velocity profile indicate the velocity field at that point with the tube diameter being proportional to the shear rate. Inlet velocities and their corresponding velocity profiles are highlighted at 0.2, 0.4 and 0.8 s, marked as ‘#’, ‘+’ and ‘*’. Arrows in the second panel indicate the direction of flow along the CA.
Table 3.
Clinical outcomes of patients.