Simulation of fluid flow through shunt
Modeling of pressure differential through the shunt was accomplished using the HPE equation in MATLAB (Mathworks, Natick, MA) by varying flow rate, shunt diameter, and shunt length. The viscosity (0.72 cP) and average flow rate (150 µL h-1) of human aqueous humor at physiological conditions were input for the values of µ and Q, respectively17.
Electrospun shunt manufacture
Electrospun shunts were manufactured utilizing a grounded collector consisting of a drill chuck (McMaster-Carr, Elmhurst, IL) and parallel collector stand positioned perpendicular to the polymer jet. Stainless steel wire (McMaster-Carr) with a diameter of 50 (PS), 75 (PCS), or 100 µm (control PS (CPS)) was inserted through a 25 G, 3.81 cm long, blunt tip needle (Nordson EFD, Westlake, OH) which was then placed into the head of the drill chuck. PET (Nanofiber Solutions, Columbus, OH) was dissolved in 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP; Sigma Aldrich, St. Louis, MO) at 10 wt% by stirring at 45 °C for 24 h. For PS, the polymer solution was then electrospun onto a 50 µm wire rotating clockwise at 175 rpm by applying 15 kV of voltage from a power supply (Glassman High Voltage, High Bridge, NJ) to a 20 G blunt tip needle (Nordson EFD) with the polymer solution flowing at a controlled rate of 1.5 mL h−1 using a syringe pump (New Era Pump Systems, Farmingdale, NY) until a 400 µm outer diameter (OD) was achieved. The CPS was manufactured similarly using a 100 µm template wire and an OD of 425 µm. For PCS, PGA was first dissolved in HFIP at 10 wt% prior to electrospinning at a flow rate of 850 µL h−1 at a voltage of 15 kV onto a rotating 75 µm template wire, until reaching an OD of 100 µm. PET was then spun around the PGA core until reaching an OD of 425 µm, as described above. All shunts were heated for 24 h at 100 °C prior to cooling to RT.
A total of 4 different shunt designs were manufactured for implanted. All implants were 6 mm in length, but varied in internal lumen diameter and design:
Open-lumen PET shunt (open PS) was fabricated from PET nanofibers with an internal lumen diameter of 50 µm.
Closed-lumen PET shunt (closed PS) has the same design and dimensions as the open PS but the internal lumen was occluded with a metal template wire
Pressure Control Shunt (PCS) was fabricated with an internal layer of PGA nanofibers surrounded by an external layer of PET nanofibers. The internal lumen expanded from 75 µm to 100 µm as the PGA degraded.
Control PS (CPS) shunt was fabricated from PET nanofibers with an internal lumen diameter of 100 µm.
Shunt OD was measured via optical microscopy using an Eclipse TS100 (Nikon Instruments, Melville, NY). Inner lumen diameter and shunt morphology were examined via SEM at 1 kV with a LEO Field Emission SEM (Zeiss, Oberkochen, Germany) following desiccation for at least 24 h and sputter coating with 10 nm of Au/Pd. Imaging was conducted prior to removal of the template wire, after removal of the wire, and following in vitro fluid flow studies. Image analysis and shunt dimensions were measured using ImageJ (U. S. National Institutes of Health, Bethesda, MD, https://imagej.nih.gov/ij/).
Evaluation of In vitro fluid flow through shunt
Experimental PS were cut to 5, 6, or 7 mm in length (n = 3 for each condition), after which the template wire was removed. The shunt was then connected in circuit to a syringe pump and manometer (NETECH, Farmingdale, NY) via the 25 G needle. Flow studies were conducted at RT. Dulbecco’s Phosphate Buffered Saline (PBS, ATCC, Manassas, VA) was pumped through the shunt at 50, 100, and 200 µL h−1. Prior to flow studies, PBS was pumped through a 25 G needle at the specified flow rate and the resulting pressure measurement was subtracted from the pressure observed with attachment of the shunt. Pressure measurements were recorded at least 30 min after a change in flow rate and only if the pressure remained constant for more than 5 min. Theoretical pressure estimates were obtained from the HPE with 25 °C PBS viscosity of 0.9 cP47. PCS (n = 3) were cut to 6 mm in length and evaluated similarly. Long-term flow studies were conducted at 37 °C by flowing PBS through PCS at 150 µL h−1 for 28 days.
In vivo performance and iop measurement
All animals in these studies were cared for in accordance with protocols approved by the Johns Hopkins University Animal Care and Use Committee, with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and with the National Institutes of Health guide for the care and use of laboratory animals. Adult NZW rabbits (Robinson Services, Mocksville, NC) were sedated by subcutaneous injection of ketamine (75 mg kg−1, Fort Dodge Animal Health, Fort Dodge, IA) with xylazine (5 mg kg−1, VedCo Inc., Saint Joseph, MO). A drop of 0.5% proparacaine hydrochloride ophthalmic solution (Bausch & Lomb, Tampa, FL) followed by a drop of 5% betadine solution (Alcon, Fort Worth, TX) was administered to the operative eye and allowed to dry for 5 min. MMC (4 mg mL−1, 50 µL) was injected into the superotemporal subconjunctival space of rabbits receiving PCS and CPS implants using a 30-guage needle. After placement of an 8–0 silk, stay suture (Ethicon, Somerville, NJ) in the superotemporal limbus, a two-clock hour, fornix-based, conjunctival peritomy was created, and the conjunctiva was dissected posteriorly. A 25 G needle (Fisher Scientific, Waltham, MA) was used to create a 2 mm long scleral tunnel prior to entering the anterior chamber. The shunt was gently guided through this tunnel. Once the shunt was in position, the template wire was removed and a 10–0 nylon cross-stitch (Ethicon) was placed at the site of tube entry to prevent fluid leakage around the tube. Balanced saline solution (BSS, Alcon) was irrigated in the anterior chamber using a 30 G needle (Fisher Scientific) to validate shunt patency. Conjunctiva was approximated to the limbus using two 10–0 nylon sutures and, again, BSS was irrigated into the anterior chamber to ensure that a water-tight bleb had formed. The stay suture was removed and topical antibiotic ointment was administered to the eye. In this manner, 3 open- and closed-lumen (containing the template wire) PS, 3 open-lumen PCS, and 4 open-lumen CPS were implanted. All shunts were 6 mm in length. IOP was measured and calibrated as previously reported48. Baseline IOP reported in Table S1 was measured in non-operated rabbits for open- and closed-lumen PS, and pre-operatively in the operated eye for PCS and CPS. IOP was measured post-operatively and at 1, 4, 7, 11, 14, 21, and 27 days for eyes containing PS and PCS using a TonoVet (iCare, Vantaa, Finland) rebound tonometer in awake, restrained rabbits without topical anesthesia as described previously49. Eyes containing a CPS implant were evaluated on day 1 post-operatively after which the rabbits were euthanized. A modified Moorfields bleb grading system was used to assess bleb size, height, and vascularity in all eyes and eyes were clinically evaluated for inflammation, infection, and anterior chamber pathology50. Shunt patency was evaluated weekly via irrigation of fluorescein (Sigma Aldrich) into the anterior chamber. 0.1% fluorescein (< 0.05 mL) was injected manually into the anterior chamber and care was taken not to elevate IOP above 25 mmHg. Fluorescein venting to the subconjunctival space was graded as either a positive (fluorescein visualized in the subconjunctival space through the shunt lumen), negative (no fluorescein), or leak (fluorescein in the subconjunctival space but not through the shunt lumen). Prior to fluorescein irrigation procedures, eyes were examined under an operative microscope for anterior chamber inflammation or fibrin formation, anterior chamber neovascularization, cornea neovascularization, cataract formation, and other signs of anterior chamber or surface pathology.
Assessment of in vivo biocompatibility
Open- and closed-lumen PS (n = 3, each), and PCS (n = 3) were implanted into NZW rabbits via an ab externo procedure, as described. After monitoring for 27 days, rabbits were euthanized and eyes enucleated, fixed in formalin, embedded in paraffin, cross-sectioned, and stained with H&E and Masson’s trichrome for histological evaluation.
Pressure and IOP measurements are presented as mean ± SE for specific time points and as mean ± SD for averages of all IOPs throughout the duration of the experiment. Shunt dimensions are also presented as mean ± SD. IOP data were analyzed using a Student’s t-test, and differences between parameters were considered statistically significant for p values < 0.05. Statistical significance of post-operative day 1 IOP values for control, PCS, and CPS eyes was determined via one-way ANOVA followed by Tukey test.