In vivo distribution of particulate matter from coated angioplasty balloon catheters
David E. Babcock a,*, Robert W. Hergenrother a, David A. Craig b, Frank D. Kolodgie c, Renu Virmani c a SurModics Inc., Eden Prairie, MN, USA b SynecorLabs LLC, Durham, NC, USA cCVPath Institute Inc., Gaithersburg, MD, USA
a r t i c l e i n f o
Article history: Received 5 December 2012 Accepted 5 January 2013 Available online 31 January 2013
Keywords: Surface modification Hydrogel Particulates Animal model Friction Hydrophilicity
a b s t r a c t
Most catheter-based vascular medical devices today have hydrophilic lubricious coatings. This study was designed to perform a territory-based downstream analysis of end organs subsequent to angioplasty with coated balloon catheters to better understand the potential in vivo physiological consequence of coating wear materials. Coronary angioplasty was performed on swine using balloon catheters modified with two polyvinylpyrrolidone (PVP)-based coatings of similar lubricity, but different levels of particu- lates (5-fold) when tested in a tortuous path model. Myocardial tissues examined 28 days post- angioplasty revealed no visible particulates in the animals treated with the lower particulate catheters while 3 of 40 sections from higher particulate catheters contained amorphous foreign material, and 1 of 40 sections from tissue treated with uncoated catheters had amorphous foreign material. Non-target organs and downstream muscle revealed no particulates for any of the treatments. Histological analy- sis showed that the overall number of vessels with embolic foreign material was low and evidence of myocyte necrosis was rare with either of the coatings investigated in this study.
! 2013 Elsevier Ltd. All rights reserved.
The use of percutaneous, catheter-based vascular devices to treat the symptoms of cardiovascular disease has become the standard of care. Many of these interventional cardiovascular devices incorporate a hydrophilic lubricious coating in order to ease movement through the vasculature. Lubricious coatings have been used for over 20 years [1,2] and the benefits are well established: (1) lower frictional force between the device and the vessel reduces tissue damage  and prevents vasospasm ; (2) improved maneuverability aids navigation of complex lesions and facilitates access to tortuous vascular sites leading to expansion of the patient population that can benefit from these treatments; and (3) reduces thrombogenicity [1,4]. In addition, reduced friction between the therapy catheters and support catheters leads to improved out- comes, reduced procedure time, and, ultimately, reduced cost .
The most commonly used lubricious coatings in medical devices consist of chemically crosslinked water-soluble polymers, such as polyvinylpyrrolidone (PVP) or polyacrylamide. When exposed to aqueous environments, these coatings form a hydrogel. Hydrogel
coatings can decrease the frictional force exerted between devices 10 to 100-fold . The enhanced lubricity derives in large part from the hydrated nature of the gel: In the swollen state, the hydrogel can be up to 90% or more water. When a swollen hydrogel is compressed, some of this water is released  and acts as a lubricant. The prop- erties of a hydrogel that impart lubricity e the ability to imbibe and exude water e also make hydrogels prone to mechanical failure . Any hydrogel-based coating, if subjected to a sufficient amount of mechanical stress, has the potential to fracture and abrade from the medical device surface.
In the past few years, regulatory agencies have increased scrutiny of coated medical devices as possible sources of foreign particulate matter to the vasculature . While there are two U.S. FDA guidance documents for the testing of percutaneous transluminal coronary angioplasty (PTCA) balloon catheters [10,11], an ASTM method  and an AAMI technical information report  for collection and analysis of particulate matter obtained from medical devices, there are no specifications regarding the levels of particulate matter gen- erated by coated medical devices. Factors that can affect device performance and resulting particulate matter generation include device materials and design, clinical technique and procedure, and device coatings.
Particulate matter in the vasculature, if large enough and in sufficient quantities, can cause occlusion of blood vessels that can
* Corresponding author. E-mail address: firstname.lastname@example.org (D.E. Babcock).
Contents lists available at SciVerse ScienceDirect
journal homepage: www.elsevier .com/locate/biomateria ls
0142-9612/$ e see front matter ! 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biomaterials.2013.01.040
Biomaterials 34 (2013) 3196e3205
lead to tissue hypoxia and, ultimately, necrosis . Microspheres made from various materials such as polyvinyl alcohol, chitosan, trisacryl gelatin, albumin, starch, poly(D,L lactide/glycolide), and polystyrene have been considered for use in therapeutic emboli- zation to treat hemorrhage or tumors . For these applications, the effective size range of microspheres is typically from 100 to 600 mm. In the field of interventional cardiology, there are numerous reports of adverse clinical events related to athero- sclerotic debris arising from sources such as acute plaque rupture or erosion [16e21]. These events often result in downstream microvascular obstruction, thrombosis, stroke, and/or myocardial scarring. Cotton fiber emboli that may originate from sterile drapes, airborne dust, or gauze are common during angiographic pro- cedures and such emboli are usually asymptomatic . A few case studies have reported clinical observation of foreign body emboli
related to polymer coated endovascular devices [23e25]. Mehta et al.  described nine cases of iatrogenic embolization of hy- drophilic polymer in patients who underwent multiple vascular interventions with hydrophilic-coated devices. However, there are no definitive preclinical references that characterize hydrogel wear particulates from coated medical devices and accompanying potential for adverse effects.
Our objective in this work was to gain a better understanding of the effects of hydrogel particulates if they are introduced to the vasculature during the use of devices containing lubricious, photo- crosslinked hydrophilic coatings. The ability to optimize lubricity while lowering particulate generation has been an ongoing industry challenge. Recent advances in coating formulations have led to the development of a submicron coating, which is designed to match the lubricity performance of traditional hydrophilic
Fig. 1. (A) Stained myocardial sections showing embolic foreign material from the apical right ventricle viewed 3 days after injection of a single swine with a coating particulate suspension. (B) Stained PVP coating coupon controls. (C) Confocal Raman spectral analysis of non-birefringent, amorphous basophilic embolic foreign material found in myocardial tissue section. Spectral match with hydrogel coating reference confirms identified foreign material is coating particulate (PVP).
D.E. Babcock et al. / Biomaterials 34 (2013) 3196e3205 3197
coatings while significantly reducing particulate generation. In this study, two coatings were investigated: one coating with an approximate dry thickness of 2 mm (hereafter referred to as “micron coating”) and a coating with an approximate dry thickness of 0.5 mm (hereafter referred to as “submicron coating”). We hypothesized that the submicron coating would generate fewer particulates than the micron coating. The primary endpoint of this study was histopathological evaluation of the 28-day time point of distal bed muscular tissues (primarily myocardium or specific skeletal muscle) known to be downstream of areas in which intra- arterial deployment of coated PTCA balloon catheters and/or injection of particulate suspensions occurred.
2. Materials and methods
2.1. Accessory devices
Cordis VISTA BRITE TIP” guiding catheters (5 and 6 French, JR4) and Hi-Torque Floppy II guide wires with MICROGLIDE coating (Abbott Vascular, 0.01400 , 190 cm) were obtained from SynecorLabs. Hemostasis valve Y-connectors (#80395) were from Qosina.
2.2. Lubricious coating formulations
Coating solutions were prepared by combining various polymers with benzophenone-based photo-reactive ionic crosslinkers in mixtures of isopropyl alcohol and water. Both the micron and submicron coatings (SurModics, Inc.) con- tained photo-reactive polyvinylpyrrolidone (PVP) and the submicron coating also contained polyacrylamide polymers that incorporated pendant benzophenone groups . The micron coating also contained unmodified pharmaceutical grade PVP (Kollidon” 90 e BASF) and was applied in one thin layer with an approximate dry thickness of 2 mm. The submicron coating did not contain unmodified PVP, and was applied as two very thin layers with a total approximate dry thickness of 0.5 mm. Approximate coating thickness was determined by cryo-sectioning coated substrates and viewing the cross-section of the coating by scanning electron microscopy.
2.3. Preparation of coated plastic rods and PTCA balloon catheters
Surrogate catheter substrates made from extruded Pebax” plastic rods (1 mm O.D., 60 cm long, shore hardness 72D e Optinova-MLE, Inc.) were used to generate particulate suspensions for intra-arterial injections. Coatings were applied to 20 cm of one end using standard dip-coating methods described previously . Briefly, the rods were wiped clean with isopropyl alcohol and allowed to air dry. The rod ends were dipped and withdrawn from coating solutions at velocities ranging from 0.3 to 0.9 cm/s and hung on racks to air dry. Dry coated rods were suspended midway between opposing UV flood-lamps and rotated during the cure to ensure even surface illumination. The curing step initiates the formation of covalent bonds between benzophenone moieties and any available abstractable hydrogen within the coating or on the substrate . The coated rods were packaged and then ster- ilized with ethylene oxide gas (EtO). Coatings were also applied in a similar manner to the distal 23 cm of PTCA catheter shaft and balloon assemblies (3! 20 mm balloon size e Creggana-Tactx Medical, Inc.). The catheters were coated while the balloons were fully inflated. The balloons were coated to maximize the amount of coating on the catheters in order to increase the potential for generating particulates during the angioplasty procedures. After coatings were complete, the balloons were deflated, pleated, folded, and repackaged prior to sterilization with EtO.
2.4. Lubricity and durability testing
The lubricity of PTCA catheter shafts was tested by vertical pinch using a DL1000 friction tester (OakRiver Technology Corporation). All samples were hydrated in phosphate-buffered saline (PBS, pH 7.4) for “1 min before testing. The catheter assembly was fixtured so that the distal coated shaft would slide up and down between two silicone rubber gripper pads that were positioned just above a beaker filled with PBS. The force of the silicone rubber gripper pads was set to 750 g. A load cell measured the force needed to withdraw the sample as it traveled (10 cm at 0.5 cm/s) through the pads. Samples were cycled 15 times. The average coefficient of friction for each cycle was calculated by dividing the load cell force by the pinch force of the gripper pads.
2.5. PTCA balloon catheter simulated-use tracking and particulate collection
Benchtop particulate suspensions were generated by tracking coated PTCA balloon catheters through a coronary model as specified in the ASTM F2394-07 method . This model is a two-dimensional simulation of a human coronary vessel without simulated lesions. The model was flushed with ISOTON” II diluent
(Beckman Coulter) prior to use. A 6 Fr guide catheter was tracked through the model up to the point where the highly tortuous path begins. A coated PTCA balloon catheter was hydrated for 1 min in diluent and then tracked over a guide wire until the balloon exited the model into a mock vessel made of silicone rubber tubing. The balloon was inflated to 15 atm, held for 30 s, deflated, and retracted through the model. The guide catheter was flushed twice with 20 ml volumes of diluent, once after catheter insertion and again after retraction. Finally, the mock vessel was flushed with 10 ml and all flush solutions were pooled in pre-cleaned 60 ml glass vials. Samples were allowed to degas by standing for 1 min before analysis by light obscuration (LO).
2.6. Particulate analysis by light obscuration
A HACH HIAC 9703þ instrument equipped with an HRLD400 sensor and PharmSpec 3.0 software was used for measuring particulate counts and sizes. The instrument was used in Run Counter mode, with n¼ 4 sample aliquots of 5 ml each analyzed per suspension. The first run was discarded, as per United States Phar- macopeial Convention Standard 788 Particulate Matter in Injections (USP <788>) , and particulate counts were averaged for the subsequent three runs and reported cumulatively (“10 mm, “25 mm, “50 mm, and “100 mm).
2.7. Coating particulate suspensions for intra-arterial injections
Particulate suspensions for intra-arterial administration were prepared imme- diately prior to injection. A hemostasis valve was attached to a 6 Fr guide catheter that had been shortened to 30 cm length. The guide catheter was inserted into the sterilized model fixture that had been flushed with normal saline. Sterile, coated Pebax” rods were hydrated for 1 min in normal saline and inserted into the guide catheter until the distal portion exited the model. The guide catheter was flushed with 7.5 ml of normal saline and the particulates were collected in a sterile, pre- cleaned glass vial. The test rod was removed and the model was flushed with another 7.5 ml that was collected and pooled with the first. This process was repeated with three additional rods and the four flush samples were pooled (a total of 60 ml). Particulate suspensions were inverted several times before withdrawing a sample for injection to ensure any larger particulates had not settled out. A sample of each particulate suspension was reserved and analyzed by LO to determine the number and size distribution of particulates.
2.8. Animal care and preparation
Thirteen healthy non-atherosclerotic domestic Yorkshire crossbred swine (juveniles, 30e50 kg) were used. Study protocols were in compliance with the National Research Councils “Guide for the Care and Use of Laboratory Animals” (8th Edition, 2011) and were approved by the testing facility’s Institutional Animal Care and Use Committee. All animals received dual anti-platelet therapy comprising