Successful use of an In Vivo Model of Spontaneous Angiogenesis as a Vascular Interface for an Implantable Bioartificial Hemofilter

The authors, as well as others, have previously shown that models of in vivo spontaneous angiogenesis are an ideal platform for the engineering of three-dimensional tissue constructs. In this reported experiment, they sought to investigate the ability of one such model to act as a vascular interface for the collection of ultrafiltrate to replace glomerular function.

The vascular pedicle of the femoral artery and vein were isolated in silicone chambers in the groins of nine rats. As previously shown, this induces robust angiogenesis, filling the chambers. An implantable device was fabricated to interface with the newly-formed vascular bed. The first part incorporated hollow collecting fibers (HFs), which have specific permeable selectivity. The second part consisted of an artificial bladder and a port for withdrawing accumulated fluid. HFs with nominal molecular weight cut-offs of 50 kD and 100 kD were used in two groups. In a third group, 100 kd HFs were used, and cultured renal tissue was transplanted into the chambers around the vessels and HFs. Three devices in each group were implanted. Fluid was tapped from the ports three times per week over a 6-week period. Ultrafiltrate volumes, blood urea nitrogen (BUN), creatinine (Cr), albumin and total protein were measured. After 6 weeks, the chambers were harvested, and the interface between the newly formed vascular bed and the collecting fibers was examined morphologically.

Angiogenesis was assessed in the implants quantitatively by vascular density (VD) at 6 weeks and averaged 158, 308, and 129 (vessels/mm², p < 0.02) for the three groups, respectively. VD correlated to percent vascular cross section (p < 0.002). All devices generated ultrafiltrate with protein permselective properties with BUN and Cr concentrations similar to plasma levels. Filtrate to plasma protein ratios were found to be less than 0.5 in all groups. Significant differences in the three groups (p < 0.05) for total ultrafiltrate volume were also observed.

Using a model of spontaneous angiogenesis in the rat, the authors showed the feasibility of creating a vascular-to-collecting-system interface. This has led to the development of a completely implantable bioartificial hemofilter with the capability to actively produce ultrafiltrate. As the first renal transplant was performed by a plastic surgeon, and microsurgeons have first-hand knowledge of angiogenesis, the authors thought that ASRM was an ideal forum for the introduction of the new technique.