Biomaterials 312010 53725384
`
`Contents lists available at ScienceDirect
`
`Biomaterials
`
`ELSEVIFP
`
`journal homepage wwwelseviercomlocatebiomaterials
`
`Delivery of paclitaxel from cobaltchromium alloy surfaces without
`polymeric carriers
`Gopinath Mania Celia E Macias a Marc D Feldman ab Denes Marton ad Sunho Oh
`C Mauli Agrawal a
`
`a Department of Biomedical Engineering College of Engineering The University of Texas at San Antonio One UTSA Circle San Antonio Tx 782490619
`USA
`b
`Division of Cardiology Department of Medicine The University of Texas Health Science Center at San Antonio 7703 Floyd Curl Drive San Antonio TX 782293900
`The Department of Veterans Affairs South Texas Health Care System San Antonio Tx 782294404
`USA
`Department of Radiology School of Medicine
`The University of Texas Health Science Center at San Antonio
`
`7703 Floyd Curl Drive San Antonio Tx 782293900
`
`d
`
`USA
`
`USA
`
`ARTICLE INFO
`
`ABSTRACT
`
`Polymer based carriers are commonly used to deliver drugs from stents However adverse responses to
`polymer coatings have raised serious concerns This research is focused on delivering drugs from stents
`without using polymers or any carriers Paclitaxel PAT an antirestenotic drug has strong adhesion
`towards a variety of material surfaces In this study we have utilized such natural adhesion property of
`to attach these molecules directly to cobaltchromium CoCr alloy an ultrathin stent strut
`PAT
`material Four different groups of drug coated specimens were prepared by directly adding PAT to CoCr
`alloy surfaces Group A PAT coated unheated and ethanol cleaned Group B PAT coated heat treated
`and ethanol cleaned Group C PAT coated unheated
`and not ethanol cleaned and Group D PAT
`coated heat treated and not ethanol cleaned In vitro drug release of these specimens was investigated
`using high performance liquid chromatography Groups A and B showed sustained PAT release for up to
`56 days A simple ethanol cleaning procedure after PAT deposition can remove the loosely bound drug
`crystals from the alloy surfaces and thereby allowing the remaining strongly bound drug molecules to be
`released at a sustained rate The heat treatment after PAT coating further improved the stability of PAT on
`CoCr alloy and allowed the drug to be delivered at a much slower
`7
`rate especially during the initial
`days The specimens which were not cleaned in ethanol Groups C and D showed burst
`release PAT
`coated CoCr alloy specimens were thoroughly
`using scanning electron microscopy
`and Xray photoelectron
`These techniques were collectively
`atomic force microscopy
`spectroscopy
`the morphology distribution and attachment of PAT molecules on CoCr alloy
`in studying
`from CoCr alloy surfaces
`surfaces Thus this study suggests
`the potential
`for delivering
`paclitaxel
`
`characterized
`
`useful
`
`without using any carriers
`
`0 2010 Elsevier Ltd All rights reserved
`
`Article history
`Received
`5 February 2010
`Accepted 16 March 2010
`Available online 15 April 2010
`
`Keywords
`
`Drug eluting stents
`
`Drug delivery
`CobaltChromium alloy
`Surface treatment
`
`Surface modification
`
`1 Introduction
`
`Atherosclerotic
`
`disease of
`
`the coronary arteries can lead to
`
`infarction heart attack 1 Atherosclerosis can be
`myocardial
`treated using implantation of metal stents which opens up the
`artery and restores blood flow to the heart 23 However stent
`deployment often involves damage to the arterial wall and causes
`secondary neointimal hyperplasia 4 Neointimal hyperplasia is
`
`primarily characterized by migration and growth of smooth muscle
`cells followed by extracellular matrix synthesis and deposition
`
`Corresponding author Tel +1 210 458 5081 fax +1 210 458 7007
`Email address GopinathManiutsaedu G Mani
`
`01429612$
`
`see front matter © 2010 Elsevier Ltd All rights reserved
`
`doi101016jbiomaterials201003043
`
`of
`
`the lumen 5 Drug eluting stents
`resulting in re occlusion
`DES which release drugs to inhibit the growth and secretion of
`smooth muscle cells are currently
`used to effectively treat
`condition 67 Although the results of shortterm clinical
`trials of
`DES demonstrate their usefulness the longterm trials showed
`risks of late stent thrombosis with adverse clinical events 810
`Polymer based drug delivery carriers in stents have been shown to
`reactions in patients 1115 Also polymers inter
`cause adverse
`and are
`fere with the blood vessels normal healing process
`considered to be a major factor in very late thrombosis 89 Hence
`several alternate techniques are under development
`drugs from stents without using polymers 1625
`The non polymer based drug delivery platforms available for
`two different
`coronary stents can generally be classified under
`
`this
`
`to deliver
`
`Abraxis EX2033
`Actavis LLC v Abraxis Bioscience LLC
`1PR201701101 1PR201701103 1PR201701104
`
`

`

`G Mani etal Biomaterials 312010 53725384
`
`5373
`
`1
`
`for
`
`1
`
`to
`
`than
`
`categories physical and chemical modification of stent surfaces
`The physical modification of stents includes porous textured and
`reservoir surfaces 26 The chemical modification includes the use
`of molecular coatings such as selfassembled monolayers 2223
`sizes of pores nanopores
`Porous
`stents with different
`100 nm micropores
`to 100 pm macropores
`greater
`100 pm have been created
`loading and releasing the drugs
`directly from stent surfaces 21 Nanoporous surfaces have been
`created by coating the stents first with a thin layer of aluminum
`conversion of aluminum to nano
`followed by electrochemical
`porous aluminum oxide 16 However
`the particle debris liberated
`reac
`from such ceramic coated stents has caused serious adverse
`tions in an animal model 27 Nanoporous metals 28 and carbon
`coatings 18 have
`carbon nanoparticle
`been previously
`also
`explored for delivering drugs directly from stent surfaces Micro
`have been created on stents by sandblasting
`textured surfaces
`treatment 17 Although the clinical performance of sandblasted
`stents is encouraging the on site loading of drugs in clinics prior to
`implantation may not be a favorable approach Reservoir
`stent
`surfaces have been created by laser cut holes on stent struts and the
`holes were subsequently filled with drugs However
`the clinical
`outcomes of such stents were not significantly different
`control groups 20 The use of selfassembled monolayers for
`delivering drugs from metal surfaces is also promising 2223 and
`under
`relevant
`for delivering
`currently
`investigation
`clinically
`amount of drugs Although a few limitations have to be overcome in
`these non polymer based DES still
`the successful
`these
`use of
`
`from their
`
`systems provide promising alternative approaches
`to currently
`available adverse reaction inducing polymer coated DES The goal
`of this study is to coat drug directly on metal surfaces without using
`for a period of time In this way once
`any carriers and to release it
`the drug is delivered the underlying metal substrates may not
`create adverse
`responses The tolerability of metals in coronary
`arteries with no serious adverse
`reactions has long been estab
`
`lished 2932
`Some therapeutic drugs naturally have strong adhesion towards
`certain material surfaces 3336 Paclitaxel PAT is one such drug
`restenosis 37
`which is also commonly used for treating instent
`PAT has been shown to strongly adsorb onto different materials
`including glass 35 polypropylene 35 silicones 35 and poly
`tetrafluoroethylene 38 In this study we have explored the use of
`these molecules directly on
`strong adhesion property of PAT to coat
`CoCr alloy a material which is used for making ultrathin stent
`struts and have investigated the in vitro drug release profiles of this
`system for up to 56 days
`
`2 Materials and methods
`
`The CoCrWNi alloy substrate L605 grade was purchased from High Tech
`Metals Inc Sylmar CA Absolute ethanol 200 proof methylene chloride acetone
`and methanol were all purchased from PharmoAAPER USA and used as received
`Paclitaxel >99 purity was purchased from ChemieTek Indianapolis USA HPLC
`and methanol were purchased from Sigma Aldrich USA
`grade water acetonitrile
`
`21 Preparation of cobaltchromium alloy specimens
`
`The CoCr alloy substrate was cut into 1 cm X 1 cm specimens
`The specimens
`twice for ten minutes
`were cleaned by sonication in ethanol acetone and methanol
`each Fresh solvents were used each time The specimens were then dried using N2
`gas Thus chemically cleaned CoCr alloy specimens are referred to here as control
`CoCr
`
`22 Direct coating of paclitaxel on CoCr alloy
`
`for 1 minute to remove the
`by air drying Thus prepared
`referred
`
`to here as Group A
`
`specimens were then sonicated in 2 mL of ethanol
`weakly bound drug from the alloy surfaces followed
`coated alloy specimens are henceforth
`paclitaxel
`Another
`coated alloy specimens was prepared according to the
`set of paclitaxel
`method described above but were transferred
`to an oven for heat treatment before
`ethanol
`The specimens were heated in air at 120 ° C for 12 hours The
`cleaning
`for 1 hour before cleaning them in
`specimens were then cooled in ambient air
`ethanol
`for 1 minute under sonication followed by air drying Thus prepared spec
`imens are referred to here as Group B The only difference between Group A and
`Group B was the introduction of heating step in Group B before ethanol cleaning
`Control specimens for both Groups A and B were also prepared and are referred to
`here as Groups C and D respectively For Groups C and D the alloy specimens were
`subjected to drug deposition protocols described for Groups A and B respectively
`cleaning procedure A schematic of the
`but without employing the final ethanol
`of specimens of Groups A B C and D are shown in Fig I The
`preparation
`of specimens of Groups A to D are provided in detail
`
`description
`
`in Table I
`
`23 Drug elution studies
`Specimens 4 each of Groups A B C and D were incubated in 2 mL of phos
`phate buffered saline with 005 tween20 PBST20 pH = 74 at 37 C Tween20
`was added to increase the
`solubility of paclitaxel
`respective time
`in PBS 1391 After
`points 1 3 5 7 14 28 and 56 days the alloy specimens were taken out and
`to 2 mL of fresh PBST20
`solution The PBST20
`immediately transferred
`samples
`collected at different time points were then analyzed for the quantity of drug
`1 mL of ethanol was added to PBST
`released using HPLC It
`is important to note that
`20 sample containers polypropylene tubes and vigorously shaken before analyzing
`the samples in HPLC This was done in order to avoid any potential
`drug loss to
`surfaces as reported in other studies 1351 Alcohol based
`container
`polypropylene
`used in the literature to extract
`the drug bound to the
`washing was previously
`container
`surfaces 1361 Besides the addition of ethanol
`further increases the solu
`in PBST20 and improves accuracy for the quantification
`bility of paclitaxel
`eluted using HPLC
`
`of drug
`
`231 Highperformance
`liquid chromatography
`The HPLC analysis was carried out using a Waters 2695 separations module with
`A Nova Pak®
`Waters
`2487 Dual
`A Absorbance
`Detector
`C18
`column
`
`39 mm X 150 mm particle size 4 um Part WAT086344 Serial 111937081140
`of water and acetonitrile 4555 vv was
`77 was used A mobile phase composition
`used at a flow rate of 10 mLmin A 10 iL volume of the sample was injected for
`separations were carried out at 35 C The detector
`analysis and chromatographic
`wavelength was fixed at 227 nm
`A standard stock
`solution of Paclitaxel was prepared in absolute ethanol 200
`proof at a concentration of 1 mgmL This solution was further diluted with ethanol
`range 1 ngmL to
`and working
`solutions were prepared over a wide concentration
`1099 ngmL Standard calibration curves were obtained for working solutions and the
`curve s were linear over the ranges of 199 ngmL and 991099 ngmL with correlation
`coefficients of R2 = 09996 and R2 = 09677 respectively
`data
`The chromatographic
`collected were analyzed using a Waters Millennium 32 software data system The
`calibration graphs were constructed by plotting the concentration of paclitaxel ngmL
`on the xaxis and the area of the peak microvoltssec on the yaxis
`
`24 Surface characterization
`Specimens of Groups A B C and D were characterized
`using Xray photoelec
`tron spectroscopy XPS scanning electron microscopy SEM and atomic
`force
`microscopy AFM
`
`241 Xray photoelectron spectroscopy
`XPS measurements were performed on a Kratos Axis Ultra instrument using
`Al Ka Xray source E = 14867 eV 225 W a hemispherical
`a monochromatized
`
`and a channeltron
`detector array Survey spectra were
`electron energy analyzer
`recorded with a pass energy of 160 eV while the high resolution spectra were
`recorded with a pass energy of 20 eV for 0 Is and C Is spectra or 40 eV for Co 2 p
`Cr 2 p W 4f Ni 2 p and N Is spectra All measurements were carried out using
`a nominal photoelectron
`to the sample surface The
`takeoff angle of 90° with respect
`binding energy BE values for CoCr alloy specimens before and after the coating of
`the standard hydrocarbon C Is peak at 285 eV
`paclitaxel were corrected by setting
`The spectral deconvolution
`and peak area analysis were carried out using the Casa
`XPS software system to determine the elemental and component compositions The
`atomic
`BE values
`compositions and molar percentage
`reported
`percentage
`concentrations
`the average of three distinct
`spots on each
`of components
`represent
`standard deviations
`sample along with the corresponding
`
`A solution of paclitaxel was prepared in ethanol at a concentration of 1 mg1 mL
`A 25 uL aliquot of the prepared solution was placed on the surface of each chemically
`cleaned CoCr alloy specimen using a micropipette and the specimens were stored
`under ambient laboratory conditions The ethanol was allowed to evaporate in air
`The
`a residue of paclitaxel
`for 3 hours leaving behind
`on the alloy specimens
`
`242 Scanning electron microscopy
`A ZEISS EVO40 Germany model SEM instrument was used in this study The
`images were acquired using a working distance of 3537 mm and an accelerating
`voltage of 5 kV Prior to SEM imaging the specimens were sputter coated with a thin
`layer of carbon
`
`

`

`5374
`
`G Mani et al Biomaterials 312010 53725384
`
`Chemically cleaned CoCr
`
`Micropipette
`
`Paclitaxel
`
`in Ethanol
`1 mgmL
`
`Chemically cleaned
`CoCr
`
`25 ill micropipette
`
`25 ktl ethanol solution
`of Paclitaxel
`
`Ethanol evaporates
`room temperature
`
`3 hours
`
`1
`
`Paclitaxel deposit
`after ethanol evaporation
`
`CoCr
`
`No heat treated
`
`Ethanol cleaned
`
`Heat treated
`
`Ethanol cleaned
`
`No heat treated
`Not ethanol cleaned
`
`Heat treated
`
`Not Ethanol cleaned
`
`Lf
`
`Group A
`
`Group C
`Fig 1 Schematic of preparation of paclitaxel coated cobaltchromium alloy specimens of Group A B C and D
`
`Group B
`
`Group D
`
`243 Atomic force microscopy
`All AFM images were
`instrument Digital
`acquired with a Nanoscope III
`Instruments Inc Santa Barbara CA at room temperature
`in air The images were
`captured in tapping mode using Si3N4
`cantilevers with a spring constant of
`2080 Nm All the reported images were flattened using a third degree polynomial
`fit The rootmean square RMS roughness values were calculated from the average
`of at least four distinct spots 10 pm x 10 pm on two different samples and reported
`standard deviations
`here with their corresponding
`
`Table 1
`
`Descriptions of control CoCr and paclitaxel coated CoCr specimens of Groups A
`B C and D
`
`Sample Name
`Control CoCr
`
`Description
`
`Chemically
`
`cleaned cobaltchromium alloy specimens
`
`Paclitaxel coated CoCr alloy
`
`Sample Name
`
`Group A
`Group B
`Group C
`Group D
`
`Paclitaxel coated
`25 ligcm2
`
`Heat treated
`
`120 °C 12 hours
`
`Ethanol
`
`cleaned
`
`Yes
`
`Yes
`
`Yes
`
`Yes
`
`No
`
`Yes
`No
`
`Yes
`
`Yes
`
`Yes
`No
`No
`
`25 Statistical
`
`analysis
`
`data collected are presented as the mean
`The experimental
`tion A oneway analysis of variance ANOVA was performed
`significance for difference was defined as p < 005
`
`standard devia
`
`and statistical
`
`3 Results
`
`31 Delivery of paclitaxel from CoCr alloy
`
`The initial drug loaded on CoCr alloy specimens of Groups A B
`C and D was 25 ugcm2 After drug deposition the specimens of
`Groups A and B were cleaned
`in ethanol
`to remove the weakly
`bound drug from the alloy surfaces The ethanol solution used for
`cleaning was then analyzed in HPLC to calculate the amount of drug
`washed out during the cleaning protocols Table 2 The difference
`drug loaded 25 ug and the amount of drug
`between the initial
`washed out
`in ethanol provided the amount of drug that actually
`retained on the alloy surface after ethanol cleaning or before
`immersion in PBST20 media Approximately 20 and 25 of the
`initial drug loaded was retained on the alloy surfaces after ethanol
`
`

`

`G Mani etal Biomaterials 312010 53725384
`
`5375
`
`0 Group A
`sGroupB
`
`a
`
`400
`
`350
`
`300
`
`a
`
`11
`
`cs7 250
`E
`
`_
`°
`1_ T
`
`10
`
`20
`
`30
`
`40
`
`60
`
`60
`
`Days
`
`Group A
`Group B
`
`10
`
`20
`
`30
`
`40
`
`50
`
`60
`
`200
`
`150
`
`100
`
`50
`
`0
`
`8 7 6
`
`4 3 2 1 0
`
``1
`
`133
`
`a
`13
`
`_
`
`O 6
`wO cn
`110
`
`702
`
`0
`
`Days
`of specimens of Groups A and B b Percentage
`Fig 2 a In vitro drug release
`of total drug release profiles of specimens of Groups A and B
`
`profiles
`
`This suggests that once the loosely bound PAT is removed by solvent
`cleaning as in Groups A and B or burst released as in Groups C and
`D the remaining strongly bound PAT molecules will be slowly
`released over a period of time
`The control CoCr alloy and four groups A B C and D of drug
`coated specimens were thoroughly characterized using SEM AFM
`and XPS to study the morphology distribution and attachment of
`PAT on CoCr alloy specimens
`
`32 SEM characterization
`
`Fig 5af shows the SEM images of control CoCr alloy and PAT
`coated CoCr alloy specimens A flat
`surface was observed
`control CoCr alloy with few surface pits Two different
`forms of
`paclitaxel crystals spheres Fig 5b and e and needles Fig Sc
`were formed on the specimens of Groups C and D The SEM images
`in determining the shape and distribution of
`are particularly useful
`aggregates of PAT molecules crystals on the specimens of Groups
`C and D Fig 5b c and e However
`the resolution of SEM limited
`imaging the molecular distribution of PAT molecules on the spec
`imens Fig 5d and f
`
`for
`
`33 AFM characterization
`
`The AFM tapping mode height and amplitude images of control
`and PAT coated CoCr alloy specimens are shown in Fig 6aj The
`
`Table 2
`Amount of drug retained on CoCr alloy specimens of Groups A and B after ethanol
`cleaning
`
`in
`
`Initial drug loaded
`Amount of drug washed out
`ethanol cleaning procedure
`Total amount of drug retained on
`the alloy surface after ethanol cleaning
`Amount of
`Initial drug loaded
`drug washed out
`in ethanol
`Percentage of total amount of
`drug washed out
`in ethanol cleaning
`Percentage of total amount of
`drug loaded on the alloy surface
`after ethanol cleaning
`
`Group
`
`25 jig
`202
`
`A
`
`07 jig
`
`Group
`
`B
`
`25 jig
`188 L03 jig
`
`48
`
`07 jig
`
`62
`
`03 jig
`
`81 3
`
`1913
`
`751 L12
`
`249 12
`
`cleaning of specimens of Groups A and B respectively The amount
`of drug that was
`retained on the specimens of Group B was
`significantly p <005 greater than that of Group A This suggests
`the heating treatment
`improves the stability of paclitaxel on
`that
`CoCr alloy surfaces
`Fig 2a shows the drug release profiles of Groups A and B in PBS
`T20 solution A sustained release profile was observed
`for both
`Groups A and B for up to 56 days The total amount of drug released
`significant between all consecutive
`was statistically
`time points
`from day 1 to day 56 for both Groups A and B Fig 2a Fig 3a
`the actual amount of drug released
`between every two
`shows
`consecutive
`time points and clearly demonstrates that paclitaxel
`was released from CoCr alloy surfaces in a sustained manner The
`actual amount of drug released
`from specimens of Group B was
`significantly lower than that of the specimens of Group A on day1
`and between days 3 and 5 Fig 3a This led to the release of less
`total drug from the specimens of Group B when compared to that of
`Group A for up to 7 days Fig 2a No statistically
`significant
`difference in the total amount of drug released was observed
`between Groups A and B from day 14 to day56 This suggests that
`a simple heating protocol provides improved stability of PAT on
`CoCr alloy surfaces This effect
`is especially significant during the
`immersion in PBST20 which cause PAT to be
`days of
`delivered from heated specimens Group B at a slower
`rate than
`unheated specimens Group A Fig 2b shows
`the percentage
`total drug release profiles for Groups A and B Approximately 7
`and 5 of the total drug retained after ethanol cleaning
`was
`from Groups A and B respectively for up to 56 days This
`suggests that PAT strongly adhered to CoCr alloy surfaces and only
`a limited amount of the total drug retained can be released for the
`time period investigated in this study
`Fig 4a shows the drug release profiles of Groups C and D PAT
`coated CoCr specimens which were not cleaned in ethanol
`in PBS
`T20 solution Burst release profiles were observed for both Groups
`C and D No significant difference in the total amount of drug
`from day 1 to day56 Approximately
`was observed
`7080 of the total drug loaded was released within day 1 and 90
`of the total drug was released in 56 days It
`is interesting to note that
`10 of the total drug loaded was retained on the alloy surface even
`of drug retained on the surface was
`after 56 days The percentage
`based on the difference between the total drug loaded
`calculated
`and the amount of drug released
`for up to 56 days The actual
`between every two consecutive
`amount of drug released
`time
`points of specimens of Groups C and D are shown in Fig 4b A
`significant amount of drug was burst released within the first 3 days
`followed by a slow and sustained release for up to 56 days Inter
`estingly the actual amounts of drug released from the specimens of
`Groups C and D after 7 days are in the same range as the amount
`from Groups A and B 48 nanogramsday
`of drug released
`
`initial
`
`released
`
`released
`
`of
`
`

`

`G Mani et al Biomaterials 312010 53725384
`
`pGroupA
`r Group B
`
`a
`
`sGroupC
`NGroupD
`
`25
`
`20
`
`2
`
`a o
`
`_
`
`ol 15
`E
`o u
`E
`
`10
`
`Nb45
`4 <
`co°
`k
`
`ttlwfi°
`69
`
`b0
`
`e°
`
`t
`
`5
`
`0
`
`10
`
`20
`
`30
`
`Days
`
`40
`
`50
`
`60
`
`Group C
`0GroupD
`
`100
`
`90
`
`ra
`
`a
`
`n0
`
`3
`
`25
`
`4
`
`ec
`13
`
`016
`46 6
`
`eiP6
`
`eep
`
`°36
`
`ezzi
`°fob
`
`e 6
`
`oc°
`
`5376
`
`a
`
`140
`
`120
`
`3
`
`41c
`
`100
`
`2_ 5 80
`o
`
`60
`
`_
`
`o 58
`
`2
`
`40
`
`_9
`
`20
`
`25
`
`ENT
`
`16
`
`20
`
`10
`
`20
`
`30
`
`46
`
`50
`
`50
`
`F
`
`13
`
`2
`
`o
`
`ac
`
`_
`c
`
`80
`
`70
`
`60
`
`50
`
`40
`
`30
`
`20
`
`10
`
`Os
`0
`
`9
`
`a
`
`g
`
`f a_
`
`o
`
`a g
`
`0
`
`2
`
`15
`
`1
`
`05
`
`_ Group C
`Group D
`
`6
`
`4a4f
`po b
`Lev
`ose
`
`aeo
`
`3 4466
`
`cri
`
`E 15
`00
`
`g
`
`zz 10
`
`5
`
`0
`
`45 0
`
`c u
`
`e5A
`
`ke 671 t
`
`Ncs
`
`be
`
`6c
`
`Nb`
`
`rs
`
`Fig 3 Amount of drug eluted between every two consecutive time points of Groups A
`and B a and Groups C and D b
`
`control CoCr showed flat
`topography with few pits and surface
`defects Fig 6a and b The RMS roughness value of control CoCr
`was 196 ± 73 nm The formation of sphere and needle shaped PAT
`crystals on CoCr was clearly observed in Groups C and D Fig 6c d
`g and h The RMS roughness values of the specimens of Groups C
`and D were 2823 ± 527 nm and 1486 ± 457 nm respectively This
`increase in surface roughness values demonstrated the
`significant
`on CoCr alloy These results are
`presence of drug aggregates
`qualitatively consistent with SEM images AFM images were also
`in observing the molecular distribution of PAT on the spec
`useful
`imens of Groups A and B Fig 6e f i and j The RMS roughness
`values of the specimens of Groups A and B were 240 ± 122 nm and
`162 ± 58 nm respectively
`
`34 XPS characterization
`
`The high resolution XPS C Is 0 Is N Is Co 2p Cr 2p W 4f and
`Ni 2p spectra were collected before and after the deposition of PAT
`on CoCr alloys The chemical compositions of control CoCr alloy
`and PAT coated CoCr alloys Groups A and C are shown in Fig 7a
`When compared to control CoCr the specimens of Group C
`showed a significant
`increase in the concentration of carbon and
`a significant decrease
`in the concentrations of oxygen and other
`
`Days
`Fig 4 a In vitro drug release profiles of specimens of Groups C and D b Percentage
`of total drug release profiles of Group C and D
`
`elements Co Cr W and Ni specific to the alloy substrate Since the
`
`PAT contains 47 carbon atoms the concentration
`of carbon is
`expected to increase after the drug deposition Also when the drug
`is deposited on the alloy the concentrations of oxygen the main
`in the surface oxides of CoCr alloy and other elements in
`element
`the alloy substrate are expected to decrease due to attenuation of
`XPS signals Such increases
`in the concentration of carbon and
`decreases in the concentrations of oxygen and other elements are
`less in Group A when compared to Group C However
`is imper
`the concentration of carbon in Group A is
`ative to note that
`significantly greater than that of control CoCr and the decrease in
`the concentrations of oxygen and other elements is significantly
`lower than that of control CoCr This strongly suggests that after
`ethanol cleaning some PAT molecules are still
`strongly adsorbed
`onto CoCr alloy surfaces Similar results were also observed
`for
`specimens of Groups D and B Fig 7b
`a trace of nitrogen
`Besides the unavoidable carbon contaminant
`contaminant was also observed on control CoCr Similar contami
`nation was reported by several other research groups for metal oxide
`samples 4041 A ratio of carbon to elements in the underlying
`metal substrate has been used in the literature to precisely determine
`of organic molecules on metal substrates
`
`the adsorptiondesorption
`
`it
`
`

`

`G Mani etal Biomaterials 312010 53725384
`
`5377
`
`irk roupC spherical crystals
`
`10 umII
`
`EMT
`W0
`
`5 09 kV
`
`16
`
`nrn
`
`4thg
`
`300KX
`
`R028531f
`
`UTSA
`
`EMT
`
`5 ODIN
`
`vi0 15 5orn
`
`lisp 303 0
`
`roe Ita0c = PL 2312 LI
`
`UTSA
`
`d Group A
`
`10 µ
`
`EHT
`WD
`
`5 00 kV
`
`350tnro
`
`Hogg 100K
`P262 V
`Ede Name
`
`e Group D spherical crystals
`
`EHT
`WD
`
`500 kV
`
`355 rffn
`
`1144g
`
`3 OD K
`Fie Name RN 30 62of
`
`UTSA
`
`EHT
`
`50014V
`
`Mac
`
`300K X
`RP 47143 NI
`
`Fds ftwre
`
`WE 35 5 arm
`mages of control CoCr alloy a group C spherical crystals b group C needle crystals c group A d group D spherical crystals e group B f
`
`UTSA
`
`IQ I
`
`EHT
`
`SCOW
`
`413
`
`37 0
`
`OCIKX
`
`FI103r
`
`HTer RP44 01
`
`UTSA
`
`Fig 5 SEM
`
`for all
`
`4245 We used such an approach
`to determine the adsorption of
`PAT on CoCr alloys Since ample amount of carbon and oxygen
`the ratios of the concentration of
`atoms are present
`in paclitaxel
`these elements to the concentrations of cobalt chromium tungsten
`and nickel were calculated to qualitatively determine the presence of
`PAT on CoCr alloys A ratio of carbon to oxygen was also calculated
`the specimens Fig 8a shows CCo + Cr + W Ni 0
`Co + Cr + W Ni and CO ratios of control CoCr and Groups A and
`C All the three ratios calculated for Group A were greater than that of
`control and lower than that of Group C This further confirms the
`presence of strongly bound PAT molecules on CoCr alloy after
`ethanol cleaning As expected the ratios were highest for Group C
`due to the presence of aggregates of drug Similar results were also
`observed for the specimens of Groups B and D Fig 8b
`The high resolution XPS C Is spectrum of control CoCr alloy
`the peaks at 285 eV
`was deconvoluted into three components
`2861 eV and 2886 eV were assigned to CC C0 and C=0 bonds
`of hydrocarbon contaminants Fig 9a and Table 3a Several studies
`in the literature have reported the adsorption of such contaminants
`on metal oxide surfaces 4546 Also the C Is peak of hydrocarbon
`
`contamination
`
`used
`
`as an internal
`
`standard for
`
`is commonly
`calibration 46 The C Is spectrum of paclitaxel in
`instrument
`form was deconvoluted into four components
`the peaks
`powder
`of the components at 285 eV 2866 eV 2892 eV and 2916 eV were
`to carbon atoms in hydrocarbon hydroxyl ester and
`assigned
`aromatic ring groups of PAT Fig 9b and Table 3a The C Is spectra
`of Groups C and D were deconvoluted into four components
`the
`peaks of the components C Is 1 C Is 2 C Is 3 and C Is 4
`were observed at 285 eV between 2864 and 2869 between 2884
`and 2889 and between 2904 and 2912 eV respectively C Is 1
`was assigned to CC bonds of PAT C Is 2 was assigned to COH
`bonds of PAT C ls 3 was assigned to 0=C0 bonds of PAT and C
`Is 4 was assigned to it > it
`shakeup satellite from the aromatic
`rings of PAT Thus the C Is components of Groups C and D are in
`excellent agreement with the C Is components of PAT powder For
`
`Groups A and B the components C Is 1 C Is 2 and C Is 3
`observed were similar to that of Groups C and D However
`4 component was missing in both Groups A and B Fig 9a and
`Table 3a This could be due to the lesser amount of drug present on
`these specimens 22
`
`the C Is
`
`

`

`5378
`
`G Mani et al Biomaterials 312010 53725384
`
`a Control CoCr
`height image
`
`b Control CoCr
`amplitude image
`
`100
`
`20v
`
`=WIZ
`
`111U
`
`d GroupC
`amplitude image
`
`f Group A
`amplitude image
`
`46tt
`is 481011
`
`ri4P
`41
`IP
`Tr222
`WV
`
`°pep
`
`fir
`
`qp
`
`I W
`
`Pe
`
`LieciiIMApas
`
`2
`
`170aRtR
`
`771111
`
`iLE
`
`Fig 6 AFM images of control CoCr ab Group C cd Group A el Group D gh and Group B ij
`
`

`

`60C
`
`IV
`
`50
`
`40 r nu
`
`N
`
`I
`
`30
`
`I
`
`20
`
`I
`
`10
`
`Fri
`
`Ii
`
`1l01
`
`a
`
`Atomicconcentration0
`
`Cis
`
`01s
`
`Nis
`
`Co 2p
`
`Cr2p
`
`W 4f
`
`Ni 2p
`
`70
`
`60
`
`Control CoCr
`
`G Ma
`
`etal Biomaterials 312010 53725384
`
`5379
`
`MConirol CoCr
`
`II
`
`Group C
`
`I
`
`Group A
`
`6
`
`5
`
`
`
`44
`
`
`
`33
`
`1
`
`
`
`00
`
`6
`
`
`
`55
`
`
`
`33
`
`
`0o
`0
`
`Fr
`Frj
`
`R
`
`L
`
`3
`
`increase
`
`properties
`
`towards different
`
`the risk of thrombosis and c delayed and incomplete
`is frequently observed 94749 Hence there is
`endothelialization
`a need to deliver drugs from stents without using polymer carriers
`In this study PAT was directly coated on CoCr alloy surfaces using
`its own strong adhesion property and the in vitro drug release
`profiles of this system were investigated
`Paclitaxel has strong binding
`material substrates such as glass polypropylene and silicone 35
`Song et al 35 examined the binding of PAT to glass and plastic
`containers in aqueous solution and tissue culture medium PAT was
`more strongly adsorbed on to glass than plastic containers Several
`other studies have also reported the adsorption of PAT on various
`plastic container
`surfaces during in vitro drug elution studies
`Palmgren et al 36 investigated the adsorption of various other
`drugs on plastic containers in deionized water and buffer solution
`Acidic drugs such as hydrochlorothiazide
`naproxen probenicid
`and indomethacin showed no adsorption on plastic containers in
`either water or buffer solution However
`basic drugs such as
`metoprolol medetomidine propranolol and midazolam strongly
`tubes especially when the storage
`adsorbed on polystyrene
`
`Control CoCr
`Group C
`Group A
`
`CCo+Cr+W+Ni
`
`0Co+CrFW+Ni
`
`CIO
`
`Control CoCr
`Group D
`Group B
`
`20
`
`18
`
`16
`
`14
`
`12
`
`10
`
`8 6 4 2
`
`ll
`
`0 r
`
`0 0 1
`
`0
`
`10
`
`10
`
`O 1
`
`0 0
`
`a
`
`ratios
`
`XPS
`
`cs
`
`771
`
`a
`
`CCo+Cr+W+Ni
`
`d0
`Fig 8 XPS determined CCo + Cr + W + Ni 0Co + Cr + W + Ni and C0 ratios of
`control Groups C and A a and Groups D and B b
`
`0Co+Cr+W+Ni
`
`Atomic
`
`concentration
`
`4
`
`3
`
`2
`
`0
`
`11 Group D
`
`Group B
`
`1
`
`I
`
`50
`
`40
`
`30
`
`20
`
`10 if
`
`Atomicconcentration
`
`Cis
`
`01s
`
`Nis
`
`W 4f
`Co 2p
`Fig 7 XPS determined atomic compositions of control Groups C and A specimens a
`and control Groups D and B specimens b
`
`Cr 2p
`
`Ni 2p
`
`The 0 Is XPS spectrum of control CoCr was deconvoluted into
`0 Is 1 at
`the peaks of
`three components
`the components
`5299 eV 0 Is 2 at 5316 eV and 0 Is 3 at 5334 eV were
`assigned to metal oxide 0i species hydroxide OH species and
`H20 species respectively Fig 9c and Table 3b The 0 Is spectrum
`form was deconvoluted into two components
`of PAT in powder
`the components at 5324 eV and 5336 eV were
`the peaks of
`assigned to oxygen atoms in OC and 0C=0 bonds respectively
`Fig 9d and Table 3b The 0 Is spectra of Groups A B C and D were
`the peaks of the compo
`all deconvoluted into three components
`nents 0 Is 1 0 Is 2 and 0 Is 3 were observed between 5299
`and 5301 between 5315 and 5322 between 533 and 5343 eV
`respectively Fig 9c and Table 3b 0 Is 1 was assigned
`to metal
`oxide species 0 Is 2 was assigned to OC bonds of PAT 0 Is 3
`to 0C=0 bonds of PAT The 0 Is components of
`was assigned
`Groups A B C and D are in good agreement with the 0 Is
`that PAT mole
`components of PAT powder These results suggest
`cules were bound to CoCr alloy surfaces
`
`4 Discussion
`
`Most currently available drug eluting stents DES in the market
`use polymerbased carriers for delivering drugs However
`there are
`a number of safety issues associated with polymerbased DES a
`hypersensitive allergic reactions 101447 b mechanical
`
`the polymers cause adverse responses including inflammatory and
`
`fissures peeling of polymer layers from the
`defects such as cracks
`have
`roughened surface with irregularities and waviness
`stent
`been observed
`in the polymer coatings of commercially available
`DES after balloon catheter expansion 48 and such defects
`can
`
`

`

`5380
`
`G Mani et al Biomaterials 312010 53725384
`
`a
`
`XPS C is soectra
`
`2887 eV
`0C0
`
`2864 eV
`COH
`
`J
`
`285 eV
`
`C C
`
`b
`
`XPS C is spectra
`
`Paclitaxel powder form
`
`> 285 eV
`
`CC
`
`285 eV
`
`CC
`
`285 eV
`
`CC
`
`285 eV
`
`C C
`
`11 groupB
`
`2912 eV
`it 91t
`Shakeup satellite OC=O
`iv group D
`
`2889 eV
`
`2869 eV
`COH
`
`iii group A
`
`2887 eV
`0C=0
`
`2904 eV
`74e
`2884 eV
`satellite 0C=0
`Shakeu