Bioinspired intratumoral infusion port catheter improves local drug delivery in the liver

Intrahepatic infusion port (conventional multi-sidehole catheter)

Harbor design

The basic strategy was to start with a standard intravenous port (PowerPort Slim Implantable Port with 6-F Catheter, BD, Franklin Lakes, NJ) and modify it to allow intratumoral catheter placement. The subcutaneous port was used unchanged. The polyurethane port catheter is too soft to easily penetrate soft tissue, so we used a 5-F braided catheter (Impress Diagnostic Peripheral Catheter, Merit Medical, South Jordan, UT) instead. Four side holes were punched along the distal 1 cm of the catheter using a 20-G needle. A gastropexy suture anchor (Cope Gastrointestinal Suture Anchor Set, Cook Medical, Bloomington, IN) was used to anchor the catheter in the liver and prevent it from migrating with respiratory motion. Heat shrink tubing (McMaster-Carr, Elmhurst, IL) was used to connect the 5-F braid catheter to the 6-F polyurethane catheter, which was then attached to the port.

Intrahepatic port placement (CT-guided)

The Institutional Animal Care and Use Committee of the Lundquist Institute approved all research procedures. All experiments were conducted in accordance with relevant guidelines and regulations and ARRIVE guidelines were followed. All procedures and imaging were performed under general anesthesia. Euthanasia was performed by intravenous administration of an overdose of barbiturate according to AVMA guidelines.

Under CT or ultrasound guidance, an 18-G Chiba needle was advanced into the target in the liver. The needle was exchanged for a 5F braided sidehole catheter over the posterior end of an Amplatz wire. Using a wire, a gastropexy suture anchor was pushed through the 5-F catheter into the liver. The 5-F braided catheter was attached to the 6-F polyurethane catheter using shrink tubing. The 6-F polyurethane catheter was then tunneled to the port pocket where both the gastropexy suture and the catheter were attached to the port reservoir. See Fig. 6.

Side hole catheter with barbs

Catheter design and manufacturing

Using parametric CAD software (OpenSCAD), we designed the barbed side hole catheter, with an outer diameter of 8F, an inner diameter of 1 mm, a 1 cm long tapered tip, 18 side holes from 1 to 3 cm from the tip, a side hole diameter of 0.4 mm and 40° angle between the side hole channels and the main catheter channel. The parametric software allows easy modifications to the side hole geometry by changing the parameters (see Parametric Catheter Model in Supplementary Materials).

The barbed side-hole infusion catheter was 3D printed with a microstereolithography printer (microArch S240, Boston Micro Fabrication, Maynard, MA) and biocompatible BMF MED resin at a resolution of 10 micrometers. Only the catheter tip was 3D printed. The catheter tip was then attached to the end of a 6-F sheath (with the tip cut) using shrink tubing and a 0.035-inch slide wire to align the lumens. The sheath serves as a catheter and as an internal stiffener so that the catheter can be advanced through the liver into the tumor. A stainless steel dowel pin (0.8 × 10 mm, McMaster-Carr) was used to close the end hole so that infusion only occurred through the side holes.

Catheter placement in the liver (CT-guided)

Under CT or ultrasound guidance, an 18-G Chiba needle was advanced into the target in the liver. The needle was exchanged for a dilator over the posterior end of an Amplatz wire. The dilator was then exchanged via the wire for the side-hole catheter with barbs. The steel dowel pin was pushed to the tip of the catheter using the wire. The internal stiffener was then removed from the catheter.

Measurement of the pull-out force

Both catheter designs (conventional side hole with suture anchor holder and side hole with barb) were placed ex vivo in porcine liver. The maximum pull-out force (Newton) was measured with a force gauge.

Measurement of local drug delivery in gel

Two catheters were tested: 1. the barbed side-hole catheter and 2. a conventional 5-F side-hole catheter with 4 side holes/cm. Side holes with an outer diameter of 0.035 inch and an inner diameter of 0.025 inch were cut in the 5-F catheter using a catheter punch (Syneo, Angleton, TX). Both catheters have a 1 cm long tip without side holes, followed by 2 cm long side holes. The end holes were closed with steel dowel pins (0.8 × 10 mm).

Glass vials were filled with a 2 cm layer of 10% (w/v) hydroxypropylmethylcellulose (bottom), followed by a 2 cm layer of saline (top). Catheters were placed in the vials so that 1 cm side holes were in the bottom gel layer and 1 cm side holes were in the top saline layer.

Methylene blue (0.02 mg/ml) was injected into the catheters. For continuous injections, 1 ml was injected over 1 minute. For pulsatile injections, 0.5 ml was injected over 0.5 s, followed by a 59 s pause, followed by a 0.5 ml injection over 0.5 s. Methylene blue in the lower gel and upper saline layers was determined by absorption measured (see Quantitative Photography in the Supplementary Materials). Experiments were performed in triplicate and mean values ​​were compared two-sidedly T-tests.

Measurement of local drug delivery in pigs

Barbed catheters and conventional side-hole catheters were placed in the livers of three Yorkshire pigs. Iohexol (276 mg/mL) was injected into the catheters and CT of the liver was performed 30 seconds after completion of the injections. For continuous injections, 3 mL was injected over 1 minute. For pulsatile injections, 1.5 ml was injected over 0.5 s, followed by a 59 s pause, followed by a 1.5 ml injection over 0.5 s. The iohexol concentration was selected to be within the maximum Hounsfield units (HU) for the CT scanner remains. Iohexol (276 mg/mL) is 2233 HU at 120 kVp, compared to the maximum HU of 3071 (which is due to the 12-bit DICOM files that can store HU values ​​in the range of -1024 to 3071).

To quantify iohexol retention at the injection site, a 3D region of interest (ROI) was drawn around the injection site and the volume and mean HU of the ROI were measured. HU of a mixture is linearly related to the HU of the components of the mixture. So: HU_ROI = v HU_contrast + (1–v) HU_liverWhere v is the fraction of the ROI volume that contains contrast. Therefore, the contrast volume in the ROI is: Volume_ROI × (HU_ROI – HU_liver)/(HU_contrast – HU_liver).

To measure the maximum iohexol delivery to the liver when administered intravenously, 120 ml of Omnipaque 300 was administered intravenously, a multiphase CT scan of the liver was obtained, and the maximum enhancement of the liver parenchyma was measured.