A novel hemostatic delivery device for thrombin Biodegradable poly

Document Sample
A novel hemostatic delivery device for thrombin Biodegradable poly Powered By Docstoc
					A novel hemostatic delivery device for thrombin: Biodegradable
poly(D,L-lactide-co-glycolide) 50:50 microspheres

Ralf Smeets,1 * Frank Gerhards,1 * Jamal Stein,2 Rui Miguel Pereira Paz,1 Stephan Vogt,3
Christoph Pautke,4 Jochen Weitz,4 Andreas Kolk4
  Department   of   Oral and Maxillofacial Surgery, University Hospital Aachen, Aachen, Germany
  Department   of   Operative Dentistry, Periodontology and Preventive Dentistry, University Hospital Aachen, Aachen, Germany
  Department   of   Orthopaedics and Traumatology, Technical University Munich, Munich, Germany
  Department   of   Oral and Maxillofacial Surgery, Technical University Munich, Munich, Germany

Received 28 July 2010; revised 3 September 2010; accepted 7 September 2010
Published online 9 November 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.a.32970

Abstract: Topical thrombins are locally active hemostatic                  ophilization was developed. The impact of several production
agents that can be used to minimize blood loss during any                  factors of the (w1/o/w2) solvent evaporation method to opti-
surgery. The aim of this study was to design and investigate               mize thrombin encapsulation, morphology of the spheres,
a thrombin-containing biodegradable hemostyptic device                     and desired drug release profile have been investigated. The
with an optimized drug release profile to promote local blood               in vitro thrombin release was dependent on the polymer-to-
clot formation. It is effective with ongoing systemic antith-              oil phase ratio, the polymer concentration, and the type of
rombotic therapy and can be used in all types of bone-related              solvent and polymer. The porosity of the spheres and release
surgery, for example, in dental surgery. Thrombin-loaded                   rate of the active agent were enhanced by increasing the
poly(D,L-lactide-co-glycolide) microspheres were synthesized               inner aqueous w1 phase. With this study, a new biodegrad-
by means of complex (w/o/w) emulsion evaporation method.                   able hemostyptic device could be verified and established for
The resulting enzyme activity of the serine-protease thrombin              a potentially safe and locally controlled thrombin release to
was verified by the specific chromogenic substrate S-2238.                   manage postsurgical hemorrhage in patients undergoing
The thrombin release profile depended on four factors: (1)                  anticoagulant therapy. V 2010 Wiley Periodicals, Inc. J Biomed

thrombin dosage, (2) polymer concentration in the o-phase,                 Mater Res Part A: 96A: 177–185, 2011.
(3) phase quotient w1:0 in the primary emulsion, and (4) the
addition of pore-introducing agents. A collagenous sponge                  Key Words: controlled drug delivery, thrombin release,
containing thrombin-loaded microspheres by means of ly-                    hemostasis, poly(D,L-lactide-co-glycolide), microsphere

INTRODUCTION                                                               pressure, ligature, sealing of bleeding vessels, and the appli-
Any bone-exposing surgical procedure such as tooth extrac-                 cation of a tourniquet. However, these methods can be labor
tion, periodontal treatment, insertion of dental implants,1 or             intensive, add time to the operative procedure, and might
fracture treatment respective orthopedic surgery on patients               be less effective in anticoagulated patients.5 Several topical
undergoing anticoagulant treatment is associated with an                   hemostatic agents are currently available in a range of
increased risk of postoperative hemorrhage.2 The danger of                 configurations. They exert their effect in a variety of ways.
postoperative bleeding depends on the extent of surgery,                   Some improve primary hemostasis, whereas others stimu-
the anticoagulant agent, and the applied dose. It varies from              late fibrin formation or inhibit fibrinolysis.
4% to 26% with the highest incidence between the second                        Some are a preparation of a procoagulant substance in
and fourth day after dental surgery.2–4 Discontinuing anti-                combination with a vehicle such as collagen matrix. Others
coagulation avoids local postoperative hemorrhage, but may                 use a matrix to provide a template for the endogenous coag-
result in serious thromboembolic complications. Therefore,                 ulation cascade to achieve hemostasis.
spatiotemporally controlled blood clot formation is a pro-                     The most promising new local anticoagulants are those
mising approach for safe bone surgery under continuous                     selectively targeting factor Xa and thrombin.6–8 Thrombin,
anticoagulant treatment, for example, administration of                    also known as activated factor II, is a serine protease of 39
salicylate and vitamin K antagonists in case of supraventri-               kDa molecular weight and almost spherical in shape. It con-
cular arrhythmias or previous pulmonary embolism.                          stitutes two peptide chains: (A) with 49 and (B) with 259
    There are several conventional hemostatic techniques to                amino acids, which are linked to each other by a disulfide
minimize blood loss. Mechanical means include manual                       bridge between position A22 and B122. During blood

*These authors contributed equally to this work.
Correspondence to: A. Kolk; e-mail: kolk@mkg.med.tum.de
Contract grant sponsor: Interdisciplinary Center for Clinical Research ‘‘BIOMAT’’

C                                                                                                                                      177
coagulation, thrombin activates fibrinogen to become fibrin          a bleeding surface activates platelets, which adhere to collagen
(factor IIa) by cleaving two specific arginine-glycine sequen-      fibrils and degranulate, triggering platelet aggregation.
ces. The resulting products are fibrinopeptide A and B and              The objective of this proof of principle study was to
the fibrin monomer, which polymerizes into strong fibrin             design and characterize a biodegradable DDS using micro-
clots. Thrombin fragments activate specific receptors on            spheres for sustained release of these thrombin-loaded
many cells to initiate the healing process, while clots dis-       microspheres.
solve.9 Clinically, thrombin formulations are used for the
treatment of acute hemorrhage, as additives in dressings or        MATERIALS AND METHODS
as tissue glue (e.g., BeriplastV, TissuecolV, and TachoCombV).
                              R           R                  R
It is known that thrombin interacts with many cell types in        Biodegradable linear poly(D,L-lactide-co-glycolide acid) (PGLA)
both initial and late stages of tissue repair/healing.10,11        having a molar composition of a D,L-lactide:glycolide ratio of
                                                                   48:52 to 52:48 (ResomerV RG 504) and free terminal
Within this investigation, a drug delivery system (DDS) with
a thrombin release profile promoting primary healing by             carboxylic acid end groups was supplied by Boehringer Ingel-
ongoing formation of local blood clotting was supposed to          heim Pharma and used without further purification and recrys-
be generated. DDSs are designed to maintain a steady drug          tallized with ethyl acetate. Its solid texture is not soluble in
level over a certain period of time to reduce the risk of          water but in organic chlorated solvents like methylene chloride.
spatiotemporal overdose. Systemic side effects should be           By means of a (w1/o/w2) emulsion evaporation method it
minimized by local application forms, immobilization, and          could be changed into spherical texture.
additional matrices such as collagen tamponades or dress-               A typical solvent cast-evaporation method was used for
ings. However, substances with low in vivo stability, such as      the preparation of films. Di-block copolymers were blended
cytokines or coagulation factors like thrombin, can be pro-        with PLGA (RG 504), at the ratio of 50:50 (% wt/wt) and
tected against denaturation or metabolic inactivation.12–14        dissolved in methylene chloride to obtain a 1 g/10 mL solu-
Some drugs that cannot be administered by this technique           tion. The polymer solution was cast onto the surface of
could benefit from injectable resorbable micro- and nano-           presilanized Petri dish, and the solvent was slowly evapo-
spheres. Conceivable applications of such particles include        rated at room temperature and then dried under high
controlled drug delivery that is localized by drug targeting,      vacuum.
detoxication, improvement of drug stability over time, and
biotechnology, including tissue engineering.                       Synthesis of thrombin-loaded
     The polymers that have recently been described to serve       poly(D,L-lactide-co-glycolide) 50:50 microspheres
as DDS are numerous and include polyurethane (main fea-            PLA and PGA are manufactured in a catalyzed process in
ture: elasticity), polysiloxane/silicone (electrical isolation),   temperatures of 140–180 C (Fig. 1). By varying the PLA/
polymethylmethacrylate (mechanical strength), polyvinyl            PGA ratio, copolymers with different physical and mechani-
alcohol (hydrophilic, mechanical strength), polyethylene           cal properties emerge. The time of degradation depends on
(tenacity), and polyvinylpyrrolidone (stabilizing suspen-          its copolymer ratio. Copolymers with a PGA mol % between
sions).15–18 A number of synthetic polymers are used               24% and 67% are considered to be amorphous.
besides natural and modified natural polymers. Resorbable               Thrombin-containing degradable microspheres were pro-
biocompatible polymers are degraded by hydrolysis or               duced by the complex (w/o/w) emulsion evaporation
enzyme reaction or both, after incorporation. Its degradation      method.22 These microspheres had an average diameter
products are finally metabolized safely and/or excreted.            between 50 and 200 lm (Fig. 1). The thrombin preparation
Among resorbable polymers, homo- and copolymers of lactic          was dissolved in a water/glycerin emulsion and dispersed
and glycolic acids show particular clinical relevance and are      in a solution of 50:50 poly(D,L-lactide-co-glycolide) and
intensively studied synthetic resorbable polymers.16–19            methylene chloride and mixed up in an ultracentrifuge
     Polylactides (PLAs) and polyglycosides (PGA) are ali-         (Ultraturrax Basic T25 IKA, Staufen, Germany) at 16,000
phatic polyesters. PLAs are more hydrophobic than PGAs             rpm, emulgated, and frozen with a combination of isopropa-
because they carry a methylene group at their a-C-atom and         nol and lyophilization below À15 C. This process that
exist in two enantiomeric forms (D- and L-lactide). The race-      needed the whole extinction of methylene chloride was per-
mat D,L-lactide-co-glycolide is noncrystalline and conse-          formed to enhance the compatibility of the phosphate buffer
quently more readily degraded by hydrolytic processes.20           solution (PBS), which was used to measure the thrombin
     In this study, we devised thrombin-loaded poly(D,L-lactide-   release rate.
co-glycolide) 50:50 microspheres (Resomer RG 504; Boeh-                Afterward, the frozen emulsion was stirred in 15 C
ringer Ingelheim Pharma, Ingelheim, Germany) embedded in           watered polyvinyl alcohol solvent to gain a complex (w1/o/
an absorbable collagenous sponge as a potential local hemo-        w2) [internal aqueous phase (w1) of water-in-oil-in-water
static agent. Collagen-based structures were already developed     (w1/o/w2) double-emulsion globules] emulsion. The meth-
in 197021 as delivery matrixes. They hold a microfibrillar          ylene chloride was entirely removed under constant stirring
structure consisting of collagen molecules with hydrochloric       at 400 (þ40) mbar. The flocked thrombin-loaded micro-
acid noncovalently bound to some of the available amino acid       spheres were isolated by centrifugation at 2000g. The sedi-
groups. The helical structure and the large surface area it pro-   ment of the microspheres was washed three times with 30
vides are important to the hemostasis achieved5 Contact with       mL of distilled water and finally lyophilized. The thrombin

178     SMEETS ET AL.                                                       BIODEGRADABLE POLY(D,L-LACTIDE-co-GLYCOLIDE) MICROSPHERES
                                                                                                                       ORIGINAL ARTICLE

FIGURE 1. Porous microspheres with a closed-cell structure and smooth surface t1 – t4 versus microspheres with a porous, foamy structure t5 þ
t6, resulting from variations in w1:0 ratios and variant temperatures t1 – t6 (140–180 C). Black bars represent 100 lm.

(synonym: coagulation factor IIa) incorporated in the 50:50             amino acids phenylalanine, tyrosine, and tryptophan yield
microspheres was of bovine plasma origin (Merck KG,                     very little interaction with this dye.
Darmstadt, Germany) lyophilized and rated at an activity of
50 NIH units/mg [The National Institute of Health (NIH)                 Analysis
standard (Lot J) is in common use for the calibration of                The release rates of active thrombin were measured on the
commercial thrombin reagents; thus, 1 NIH unit is equiva-               basis of reaction with the chromogenic substrate S-2238
lent to 1.1–1.3 IU of thrombin]. Further supplements were               (Chromogenix-Instrumentation SpA, Milan, Italy). Thrombin
mannite, Tween 80, and calcium chloride. This primary                   is a highly specific serine protease that hydrolyzes peptide
(w1/o) emulsion was cooled down to 0 C and then emul-                  and ester bonds at the C-terminal end of L-arginine. The
gated once more in a 15 C aqueous solution of polyvinyl                chromogenic substrate cleaves p-nitroaniline, which is pho-
ethanol, forming a more complex (w1/o/w2) suspension.                   tometrically measured at 405 nm. The conversion of S-2238
Thereafter, the volatile methylene chloride phase was                   is proportional to the enzymatic activity of thrombin and is
removed from the emulsion by reduction of the atmospheric               given in NIH units; 1 NIH unit equals 0.324 6 0.073 lg ¼
pressure. This procedure synthesized the thrombin-loaded                8.85 6 2.0 pmol. One milliliter of human plasma contains
microspheres.                                                           17.5 NIH units thrombin on average. Generally, the Bradford
    The microspheres were solved in raw collagen material,              test13,23 has been adjusted to demonstrate that all substan-
which was also from bovine origin and purchased from                    ces, buffers, and degradation products as used in this
Suwelack, Skin & Health Care GmbH (Billerbeck, Germany).                experiment did not influence the test results.
This mixture was pressed by means of preformed matrices
in certain forms, for example, to match the shape of a dental           Thrombin activity in relation to pH values. The activity of
alveolus and lyophilized afterward. All other chemicals were            thrombin was measured under different conditions. Initially,
purchased from Sigma Aldrich Company (Munich, Germany)                  its activity was measured with different pH values. The deg-
at the highest available analytical grade.                              radation of the microspheres leads to lactide and glycolide
    Release of thrombin from the resulting poly(D,L-lactide-            acids, which lowers the pH of the PBS. Below pH 6, throm-
co-glycolide) 50:50 microspheres was measured in PBS at a               bin rapidly loses its enzymatic activity (Fig. 2). The enzyme
temperature of 37 C by applying the Bradford method13,23               activity of thrombin has its maximum above pH 6. The
and a thrombin-specific chromogenic substrate. Within each               acidic degradation products of poly(D,L-lactide-co-glycolide)
experiment, only thrombin formulations with identical                   50:50 microspheres lactide and glycolide acids lower the pH
charge numbers were used. Van der Waals forces and                      as they are released into PBS. pH decrease was reached by
hydrophobic interactions create a complex of proteins and               adding D,L-lactic acid. The biologically active thrombin
the anionic species of the dye that specifically binds to func-          released from degradable microspheres was calibrated by
tional macromolecular groups of arginine. On the other                  stopping the enzymatic activity at 60 s by lowering the pH
hand, amino acids such as histidine and lysine and aromatic             < 2 through the addition of 20% acetic acid (Fig. 3).

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | JAN 2011 VOL 96, ISSUE 1                                                                     179
                                                                        FIGURE 4. Influence of incorporated thrombin quantity on release
                                                                        rate and amount over a 28-day interval in poly(D,L-lactide-co-glycolide)
FIGURE 2. pH-dependant thrombin activity. The optical density of p-     50:50 microspheres with identical w:o phase ratio and polymer con-
nitroaniline, cleaved from the chromogenic substrate S-2238 by          centration. The plot shows time-dependant cumulative release of
thrombin catalyzation, was measured (p < 0.01). The pH was reduced      thrombin in PBS at 37 C.
by adding D,L-lactic acid. Below pH 6 (left of dotted line), thrombin
rapidly loses its enzymatic activity.

                                                                        (o-phase)/100 mg spheres was compared with 40 mL
Relationship between incorporated thrombin quantity                     dispersed methylenechloride (o-phase)/100 mg spheres.
and release rate. Dose dependency of the delivered
amount of thrombin incorporated into the poly(D,L-lactide-
co-glycolide) 50:50 microspheres was investigated. With                 Effect of the phase ratio in the primary emulsion. The
identical w:o phase ratio and polymer concentration, the                influence of the ratio of w1:0 phases in the complex (w1/o/
quantity of thrombin was tripled per 100 mg microspheres.               w2) emulsion evaporation method on thrombin release
Lowest thrombin dose was 7.24 mg/100 mg microspheres                    from the poly(D,L-lactide-co-glycolide) 50:50 microspheres
and respective highest amount of thrombin was 21.72 lg                  was investigated accordingly. The inner aqueous phase was
(Fig. 4). Release rates were analyzed according to 2.2.                 increased from 1 (1:40) to 5 mL (5:40). The release was
                                                                        observed for 23 days according to the description in ‘‘Analy-
                                                                        sis’’ section (Fig. 5).
Effect of volume percent of organic solvent. The influence
of the concentration of the polymer in the dispersed methyl-
ene chloride o-phase during the complex (w1/o/w2) emul-                 Effect of the polymer concentration. The influence of the
sion-evaporation process on the release of thrombin was                 water-soluble, expanding additive polyethylene glycol on
investigated according to 2.2. The cumulative release                   the thrombin release over time was examined according to
rate of thrombin within 20 mL dispersed methylenechloride               the description in ‘‘Analysis’’ section.

FIGURE 3. Calibration graph for determining biologically active
thrombin released from degradable microspheres. The enzymatic           FIGURE 5. Changing the w1:o phase ratio in the complex (w1/o/w2)
activity is stopped at 60 s by lowering the pH < 2 through the addi-    emulsion–evaporation method influences the release kinetics of
tion of 20% acidic acid.                                                thrombin in PBS buffer at 37 C.

180      SMEETS ET AL.                                                             BIODEGRADABLE POLY(D,L-LACTIDE-co-GLYCOLIDE) MICROSPHERES
                                                                                                                      ORIGINAL ARTICLE

                                                                        0.05). For the ‘‘100 mg’’ (7.24 mg) thrombin preparation, a
                                                                        release rate of 230 lg (3.7 NIH units) thrombin per 100 mg
                                                                        microspheres was found in the first 36 h, which translates
                                                                        to 65% of the total amount of incorporated thrombin. The
                                                                        300 mg (21.72 lg) thrombin preparation was only released
                                                                        to 60% of the overall incorporated amount (300 lg; 12.2
                                                                        NIH units). Over the following 10 days, the ‘‘100 mg’’ speci-
                                                                        mens delivered another 35 lg (0.5 NIH units) of thrombin,
                                                                        whereas ‘‘300 mg’’ released another 50 lg (2.7 NIH units)
                                                                        per 100 mg microspheres. The ‘‘100 mg’’-group cumulatively
                                                                        released 72% of its overall thrombin contents compared
                                                                        with 68% of the 300 mg over 23 days. In contrast to the
                                                                        differences of the initial burst data within the first 60 h,
                                                                        there was no statistical significance in the cumulatively
FIGURE 6. pH record in PBS buffer solution medium before each data
readout over 23 days. Medium was exchanged at each time to pre-
                                                                        quantity. The cumulative amount of released thrombin from
vent a further drop in enzymatic activity.                              poly(D,L-lactide-co-glycolide) 50:50 microspheres was not
                                                                        proportionally related to the total amount of initially incor-
                                                                        porated thrombin because the efficiency of encapsulation
Statistical analysis                                                    capacity is continuously reduced reciprocally to increasing
All the data were expressed as mean 6 standard deviation.               thrombin concentrations (Fig. 9). The ‘‘100 mg’’ preparation
Statistics were calculated with SPSS computer software for              of poly(D,L-lactide-co-glycolide) 50:50 microspheres cumula-
Windows (version 14.0). Data were compared by a two-                    tively releases 85% of its initial thrombin amount, whereas
tailed Student’s t-test for differences between two groups;             the ‘‘300 mg’’ preparation releases only 50% of its incorpo-
at least six replicates were analyzed in each case, and differ-         rated thrombin, although the relative thrombin mass is
ences were deemed statistically significant if p < 0.05.                 increased fivefold at the same time (0.57% vs. 0.12% by
pH-dependant thrombin activity
                                                                        Effect of volume percent of organic solvent
The measured thrombin activity data in relation to certain
                                                                        The release of thrombin over 23 days depends on the
pH values ranging between 3 and 10 are presented in Fig-
                                                                        concentration of the polymer in the dispersed methylene
ure 2. The dependency of the thrombin activity was signifi-
                                                                        chloride o-phase CH2Cl2 during the complex (w1/o/w2)
cant above a pH value of 6.2 (p < 0.05). The calibration
                                                                        emulsion-evaporation process as results are presented in
curve given in Figure 2 is the reference for calculating the
                                                                        Figure 8. The standard polymer concentration is 0.05 g/mL.
active thrombin release from the degradable microspheres.
                                                                        An increase in the o-phase from 20 to 40 mL under con-
    The degradation of the poly(D,L-lactide-co-glycolide)
                                                                        stant w1-phase volume at 2.5 mg leads to a decrease in the
results in the release of lactic and glycolic acids, which leads
                                                                        average sphere diameter from 82.4 to 58.6 lm (p < 0.01).
to a decrease in the pH value and, subsequently, to a
                                                                        The reduction of the sphere size leads to an increased num-
reduced fraction of active thrombin (Fig. 6). The daily
                                                                        ber of spheres per overall mass and, consequently, an
exchange of the PBS after each data reading point demon-
                                                                        enlargement of the overall surface area and, therefore, the
strated that this undesirable effect could be prevented.
Figure 6 shows the range of the pH value between 6.6 and
7.1 during the testing period. The time dependency of the
enzymatic activity of the released thrombin was significant
(p < 0.05). Because the thrombin activity has a half-life of
about 48–60 h, it invariably biases the measured cumulative
enzymatic activity and, therefore, also the anticipated
release profile (Fig. 7).

Relationship between incorporated thrombin
quantity and release rate
There was a significant dose dependency of the delivered
amount of thrombin (Fig. 4) being incorporated into the
poly(D,L-lactide-co-glycolide) 50:50 microspheres. With iden-
tical w:o phase ratio and polymer concentration, tripling the
quantity of thrombin per 100 mg microspheres resulted in
a significant higher initial burst within the first 2.5 days,
after which the release characteristic showed a plateau par-            FIGURE 7. Time-dependent thrombin activity in PBS at 37 C (initial
allel to that of the lower thrombin dose of 7.24 mg (p <                concentration 1 NIH unit thrombin/mL PBS).

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | JAN 2011 VOL 96, ISSUE 1                                                                   181
FIGURE 8. An increase in volume of the dispersed methylenechloride
(o-phase) with constant polymer and thrombin masses and constant     FIGURE 10. Influence of the water-soluble, expanding additive polye-
w1-phase volume accelerates the cumulative release of thrombin in    thyleneglycol on thrombin release from poly(D,L-lactide-co-glycolide)
PBS at 37 C (p < 0.01).                                             50:50 microspheres provided, while thrombin mass and w1:0 phase
                                                                     ratio remain constant. Plot shows cumulative thrombin release into
                                                                     PBS buffer at 37 C over 23 days.
thrombin loading capacity. Over 23 days, a continuous
cumulative release of 28 lg thrombin per 100 mg spheres
was noted in the 20 mL group versus 40 lg thrombin per               cumulative thrombin release into PBS at 37 C over 23 days,
100 mg spheres in the 40 mL group (p < 0.05).                        whereas thrombin mass and w1:o phase ratio remain con-
                                                                     stant (p < 0.01) (Fig. 9).

Effect of the phase ratio in the primary emulsion                    Effect of the polymer concentration
The ratio of w1:0 phases in the complex (w1/o/w2) double             Influence of the water-soluble, expanding additive polyethy-
emulsion evaporation method positively influences the                 leneglycol on the thrombin release was examined by
resulting thrombin release from the poly(D,L-lactide-co-glyco-       increasing the polymer mass from 4% to 10%, which only
lide) 50:50 microspheres. An increase in the inner aqueous           tendentially enhances the thrombin release (differences not
phase from 1 to 5 mL results in a pronounced release of              significant), whereas further enhancement of up to 20 and
thrombin, creating an initial burst effect. Within the first          30 mass % had a significant effect (Fig. 10). The porosity of
12 h, the microspheres with the increased inner aqueous              the spheres and release rate of thrombin are also increased
phase (5:40) release 20 times more thrombin (approxi-                by higher concentration of the inner aqueous w1 phase of
mately 190 lg) than microspheres with the standard 1:40              the polymer. Thrombin release from poly(D,L-lactide-co-gly-
ratio (approximately 8.8 lg). Another 65 lg of thrombin              colide) 50:50 microspheres is also accelerated by addition
was released from the former microspheres over a 4-day               of water-soluble or expanding organic additive.
period, whereas the standard microspheres with a 1:40
phase ratio delivered overall less thrombin over a longer pe-        DISCUSSION
riod of time (approximately 40 lg over 23 days). Changing            The nature of the requirement for hemostasis is a key
the w1:o concentration (1:40 vs. 5:40) provided increased            determinant in the choice of agent. Intraoral bleeding condi-
                                                                     tions are rarely dry; thus, Fibrin sealants or BioGlue, on the
                                                                     other hand, are particularly poorly suited to this application
                                                                     as they require a dry field for application. The coagulation
                                                                     status of the patient is also important because agents using
                                                                     solely collagen, cellulose, or gelatin rely on intact clotting
                                                                     mechanisms. Inevitably, they are less effective in coagulo-
                                                                     pathic patients.24 In contrast, thrombin is independent of
                                                                     the clotting mechanism and, therefore, more applicable
                                                                     when faced with coagulopathy.5
                                                                         Because patient safety is, of course, of chief importance,
                                                                     there is no doubt that topical hemostats have a chief clinical
                                                                     place and, with the increasing popularity of minimally inva-
                                                                     sive or dental surgery in anticoagulated patients, where
                                                                     hemostasis is of particular concern, this is likely to expand.
                                                                     The aim of this study was to develop and characterize a de-
FIGURE 9. Thrombin contents versus efficiency of encapsulation of     vice for sustained release of thrombin for over 20 days
poly(D,L-lactide-co-glycolide) 50:50 microspheres.                   from microspheres embedded in a collagenous sponge as a

182     SMEETS ET AL.                                                          BIODEGRADABLE POLY(D,L-LACTIDE-co-GLYCOLIDE) MICROSPHERES
                                                                                                                  ORIGINAL ARTICLE

local hemostatic. In this study, thrombin of bovine origin              glycol) to the microsphere matrix leads to an accelerated
has been used because the recombinant and human plasma-                 release of thrombin from poly(D,L-lactide-co-glycolide) 50:50
derived alternative products were not yet available at the              microspheres. The water-soluble additives are being dis-
time of the experiments in 2007.25 The authors are aware                solved or expanded in contact with water, leaving water-
of the fact that the latter thrombin formulations do not go             filled pores and canals in the microsphere matrix, which in
along with potential cross-reactive antibodies.25 Apart from            turn promote the liberation of the active agent (thrombin in
this, thrombin was not used as a stand-alone product, as                our case). Drug release from microparticles has been
collagen itself has an additional coagulative function. This            described to occur by diffusion through water-filled pores
study has been initiated under the aspects of development               and after hydrolytic degradation/ erosion of the polymer.34
and characterization of a local delivery vehicle for sustained              Finally, up to a certain amount, increasing the percent-
release of thrombin, which was a component of the com-                  age of polyethyleneglycol mass in the microsphere matrix
plete hemostatic agents.26 The pure thrombin component                  accelerated thrombin release. As indicated in Figure 10, an
itself is exchangeable without influence on any spatiotempo-             increase from 4 to 20 mass % enhances thrombin release,
ral characteristics of the device.27                                    whereas a further increase to 30 mass % yields no addi-
     Since the 1960s, linear PLAs and polyglycolides as well            tional effect. This high initial burst of the drug delivery
as their respective copolymers have been used as osteosyn-              course based on water-miscible solvents stands in good
thesis- and suture materials28–30 and, most importantly in              agreement with current published data.20,35
this context, drug delivery and targeting matrices, first                    We clearly demonstrated that poly(D,L-lactide-co-glyco-
described in 1971.31 In contrast to other more unstable                 lide) 50:50 microspheres can be used efficiently for sus-
carriers with unsafe spatiotemporal properties like calcium             tained thrombin delivery. A percentage change in mass of
phosphate,1 these materials distinguish themselves by their             the embedded thrombin, polymer concentration, w1:0-phase
favorable mechanical properties, low toxicity, excellent bio-           ratio, and addition of water-soluble polymer poly(ethylene
compatibility, and predictable degradation process.19,32 In             glycol) alters the drug content and release rate of the micro-
1967, the first US patent was filed on PLA suture materials.              spheres. Experimental work is in progress to examine the
Subsequently, the first polyglycolide suture material                    in vivo performance of thrombin-impregnated poly(D,L-lac-
(DexonV) was introduced in 1970, and the first poly(lactide-
                                                                        tide-co-glycolide) 50:50 microspheres embedded in collage-
co-glycolide) suture material (VicrylV) in 1975.30 The char-
                                                                        nous foam applied as a local hemostatic agent. The widely
acteristics of drug release by microspheres depend on the               published reports are not only directed in developing
nature of the drug, the polymer applied, the fabrication pro-           appropriate dosage formulations to produce the expected
cess, and the processing parameters. Therefore, in this                 results, but also to realize that the dosages are to be
study, variant polymer concentrations have been loaded                  obtained with high batch-to-batch repeatability.36 The char-
with different thrombin masses.                                         acteristics of drug release by microspheres depend not only
     It is generally accepted that the use of unreasonably              on the nature of the drug, the polymer used, the production
high quantities of such carriers can lead to problems of car-           method, and the processing parameters20,37 but also on the
rier toxicity, metabolism and elimination, or biodegradabil-            different pH value condition being tested. Acidic pH sur-
ity. Attributable is an increase in porosity of the inner struc-        roundings (pH 3.0–6.0) did influence rupture release from
ture secondary to an increase in volume of the encapsulated             the microspheres, governing the burst phase. Low pH cata-
aqueous w1 phase. Therefore, if the phase ratio is increased            lyzes breakage of the ester linkage of the polymer backbone,
beyond a certain limit without simultaneously increasing                enhancing polymer erosion, which is known to control the
the polymer concentration, open cell and mechanically                   secondary apparent-zero order phase.38 The morphology of
instable spheres result that ultimately lose their ability to           degrading microspheres was significantly different at variant
carry any aqueous phase (as thrombin) and specific release               pH values (Fig. 2). Between the tested pH range of 6.2–11,
properties.18 The activity of the released thrombin could               comprising the variety of basic and acid pH environments of
suffer because of various steps during the manufacturing                the human body intra- and extraorally, the microspheres
course such as high shear forces exerted during the process-            showed typical erosion behavior with surface pitting and
ing of the primary (w1/o) emulsion, denaturation, and                   with an increase in the number and size of channels with
agglomeration at the hydrophilic/hydrophobic interface of               time demonstrating autocatalysis where acidic oligomeric
the primary (w1/o) emulsion, loss of active agents to aque-             units accumulate and create a porous network within the
ous phase, and the lyophilization process before storage.               microspheres, accelerating PLGA degradation. Microspheres
The degradation of PLGA occurs in two steps: First, chain               incubated at lower pH 6.1–3.0 had smoother surfaces
cleavage as indicated by a decreasing molecular weight, and,            throughout the entire study period, with no evidence of sur-
second, polymer erosion as indicated by a loss of matrix                face pitting or channels (data not shown). In addition, these
weight.33 Depending on the molecular weight of PLGA, there              microspheres seemed to become progressively brittle with
is an induction period before mass loss or polymer erosion.             time, as evident from the presence of fragmented crystal-
High-molecular-weight polymers degrade to lower-molecu-                 lized microspheres.39
lar-weight fractions, which still remain water-insoluble.                   Thrombin delivery from PLGA or PLGA-based polymers
     The addition of solid, water-soluble additives (e.g., NaCl)        is an attractive field with innumerable opportunities for fur-
or an expanding water-soluble polymer (e.g., polyethylene-              ther research and clinical impact. Topically applied thrombin

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | JAN 2011 VOL 96, ISSUE 1                                                              183
is expected to have only local effects.25 Because the throm-                    for an involvement of Gq/G11. J Biol Chem 2003;278:28743–
boembolic risk due to discontinuing the anticoagulation                         28749.
                                                                          10.   Naldini A, Carraro F, Baldari CT, Paccani SR, Bernini C, Keherly
therapy is much higher than the local bleeding danger, con-                     MJ, Carney DH. The thrombin peptide, TP508, enhances cytokine
siderable clinical applications include permanently anticoa-                    release and activates signaling events. Peptides 2004;25:
gulated patients, undergoing surgical procedures such as                        1917–1926.
                                                                          11.   Norfleet AM, Bergmann JS, Carney DH. Thrombin peptide, TP508,
tooth extraction,40 periodontal maneuvers,41 or bone sur-
                                                                                stimulates angiogenic responses in animal models of dermal
gery.1 By locally controlled and continuous thrombin deliv-                     wound healing, in chick chorioallantoic membranes, and in cul-
ery, the likelihood of postoperative bleeding complications                     tured human aortic and microvascular endothelial cells. Gen
during dental surgery will be much lower than is the risk of                    Pharmacol 2000;35:249–254.
                                                                          12.   Barman SP, Lunsford L, Chambers P, Hedley ML. Two methods
thrombosis or embolic complication after cessation of anti-                     for quantifying DNA extracted from poly(lactide-co-glycolide)
coagulant medication before surgery.                                            microspheres. J Control Release 2000;69:337–344.
                                                                          13.   Bradford MM. A rapid and sensitive method for the quantitation
                                                                                of microgram quantities of protein utilizing the principle of pro-
                                                                                tein-dye binding. Anal Biochem 1976;72:248–254.
Thrombin is a pluripotent hemostatic factor, which pro-                   14.   Takakura Y, Nishikawa M, Yamashita F, Hashida M. Development
motes coagulation, thrombosis, and local vasoconstriction,                      of gene drug delivery systems based on pharmacokinetic studies.
which makes topical thrombin an ideal agent for the promo-                      Eur J Pharm Sci 2001;13:71–76.
                                                                          15.   Tao SL, Lubeley MW, Desai TA. Bioadhesive poly(methyl methac-
tion of hemostasis during surgical procedures. This study                       rylate) microdevices for controlled drug delivery. J Control
highlights the preparation and characterization of new                          Release 2003;88:215–228.
resorbable DDSs on the basis of nano- and microspheres,                   16.   Avgoustakis K. Pegylated poly(lactide) and poly(lactide-co-glyco-
                                                                                lide) nanoparticles: Preparation, properties and possible applica-
used for spatiotemporal controlled sustained thrombin
                                                                                tions in drug delivery. Curr Drug Deliv 2004;1:321–333.
release and drug targeting. Commercially available poly(D,L-              17.   Bala I, Hariharan S, Kumar MN. PLGA nanoparticles in drug deliv-
lactide-co-glycolide) 50:50 Resomer (RG 504) was used as                        ery: The state of the art. Crit Rev Ther Drug Carrier Syst 2004;21:
the polymer, which has a long history of biomedical applica-                    387–422.
                                                                          18.   Babu VR, Patel P, Mundargi RC, Rangaswamy V, Aminabhavi TM.
tion. Nanoparticles from this polymer were followed chrono-
                                                                                Developments in polymeric devices for oral insulin delivery.
logically from nanoparticle characterization, degradation,                      Expert Opin Drug Deliv 2008;5:403–415.
biocompatibility, and in vitro release. The in vitro tests of             19.   Fischer J, Kolk A, Wolfart S, Pautke C, Warnke PH, Plank C,
the influencing environmental factors led to an optimized                        Smeets R. Future of local bone regeneration—Protein versus gene
                                                                                therapy. J Craniomaxillofac Surg. Apr 29. [Epub ahead of print.]
controlled release of thrombin as a local bioactive coagula-              20.   Mundargi RC, Babu VR, Rangaswamy V, Patel P, Aminabhavi TM.
tion device, which will be of particular clinical interest                      Nano/micro technologies for delivering macromolecular therapeu-
because of the rising number of patients undergoing antico-                     tics using poly(D,L-lactide-co-glycolide) and its derivatives. J Con-
agulation therapy.                                                              trol Release 2008;125:193–209.
                                                                          21.   Schonauer C, Tessitore E, Barbagallo G, Albanese V, Moraci A.
                                                                                The use of local agents: Bone wax, gelatin, collagen, oxidized cel-
REFERENCES                                                                      lulose. Eur Spine J 2004;13(Suppl 1):S89–S96.
 1. Smeets R, Maciejewski O, Gerressen M, Spiekermann H, Hanisch          22.   Uchida T, Yoshida K, Goto S. Preparation and characterization of
    O, Riediger D, Blake F, Stein J, Holzle F, Kolk A. Impact of rhBMP-         polylactic acid microspheres containing water-soluble dyes using
    2 on regeneration of buccal alveolar defects during the osseointe-          a novel w/o/w emulsion solvent evaporation method. J Microen-
    gration of transgingival inserted implants. Oral Surg Oral Med              capsul 1996;13:219–228.
    Oral Pathol Oral Radiol Endod 2009;108:e3–e12.                        23.   Kruger NJ. The Bradford method for protein quantitation. Meth-
 2. Pototski M, Amenabar JM. Dental management of patients receiv-              ods Mol Biol 1994;32:9–15.
    ing anticoagulation or antiplatelet treatment. J Oral Sci 2007;49:    24.   Oz MC, Rondinone JF, Shargill NS. FloSeal matrix: New generation
    253–258.                                                                    topical hemostatic sealant. J Card Surg 2003;18:486– 493.
 3. Madrid C, Sanz M. What influence do anticoagulants have on oral        25.   Cheng CM, Meyer-Massetti C, Kayser SR. A review of three stand-
    implant therapy? A systematic review. Clin Oral Implant Res 2009;           alone topical thrombins for surgical hemostasis. Clin Ther 2009;
    20(Suppl 4):96–106.                                                         31:32–41.
 4. Evans IL, Sayers MS, Gibbons AJ, Price G, Snooks H, Sugar AW.         26.   Weaver FA, Lew W, Granke K, Yonehiro L, Delange B, Alexander
    Can warfarin be continued during dental extraction? Results of a            WA. A comparison of recombinant thrombin to bovine thrombin
    randomized controlled trial. Br J Oral Maxillofac Surg 2002;40:             as a hemostatic ancillary in patients undergoing peripheral arte-
    248–252.                                                                    rial bypass and arteriovenous graft procedures. J Vasc Surg 2008;
 5. Zwischenberger JB, Brunston RL Jr, Swann JR, Conti VR. Com-                 47:1266–1273.
    parison of two topical collagen-based hemostatic sponges during       27.   Lomax C, Traub O. Topical thrombins: Benefits and risks. Pharma-
    cardiothoracic procedures. J Invest Surg 1999;12:101–106.                   cotherapy 2009;29(7 Pt 2):8S–12S.
 6. Franchini M, Mannucci PM. A new era for anticoagulants. Eur J         28.   Nordenram A, Bang G. P.G.A—A new synthetic, absorbable
    Intern Med 2009;20:562–568.                                                 suture material. A clinical and histological investigation. Sven
 7. Scotton CJ, Krupiczojc MA, Konigshoff M, Mercer PF, Lee YC,                 Tandlak Tidskr 1972;65:431–440.
    Kaminski N, Morser J, Post JM, Maher TM, Nicholson AG, Moffatt        29.   Ewers R, Lieb-Skowron J. Bioabsorbable osteosynthesis materi-
    JD, Laurent GJ, Derian CK, Eickelberg O, Chambers RC. Increased             als. Facial Plast Surg 1990;7:206–214.
    local expression of coagulation factor X contributes to the fibrotic   30.   Bezwada RS, Jamiolkowski DD, Lee IY, Agarwal V, Persivale J,
    response in human and murine lung injury. J Clin Invest 2009;               Trenka-Benthin S, Erneta M, Suryadevara J, Yang A, Liu S. Mono-
    119:2550–2563.                                                              cryl suture, a new ultra-pliable absorbable monofilament suture.
 8. Thevenet J, Angelillo-Scherrer A, Price M, Hirt L. Coagulation fac-         Biomaterials 1995;16:1141–1148.
    tor Xa activates thrombin in ischemic neural tissue. J Neurochem      31.   Chiou WL, Riegelman S. Pharmaceutical applications of solid dis-
    2009;111:828–836.                                                           persion systems. J Pharm Sci 1971;60:1281–1302.
 9. Vogt S, Grosse R, Schultz G, Offermanns S. Receptor-dependent         32.   Aboudzadeh N, Imani M, Shokrgozar MA, Khavandi A, Javadpour
    RhoA activation in G12/G13-deficient cells: Genetic evidence                 J, Shafieyan Y, Farokhi M. Fabrication and characterization of

184      SMEETS ET AL.                                                                BIODEGRADABLE POLY(D,L-LACTIDE-co-GLYCOLIDE) MICROSPHERES
                                                                                                                             ORIGINAL ARTICLE

      poly(D,L-lactide-co-glycolide)/hydroxyapatite nanocomposite scaf-      38. Faisant N, Siepmann J, Benoit JP. PLGA-based microparticles:
      folds for bone tissue regeneration. J Biomed Mater Res A 2010;             Elucidation of mechanisms and a new, simple mathematical
      94:137–145.                                                                model quantifying drug release. Eur J Pharm Sci 2002;15:
33.   Houchin ML, Topp EM. Chemical degradation of peptides and                  355–366.
      proteins in PLGA: A review of reactions and mechanisms.                39. Li L, Schwendeman SP. Mapping neutral microclimate pH in
      J Pharm Sci 2008;97:2395–2404.                                             PLGA microspheres. J Control Release 2005;101:163–173.
34.   Rothen-Weinhold A, Besseghir K, De Zelicourt Y, Gurny R. Develop-      40. Kolk A, Pautke C, Hall D, Wagenpfeil S, Wolff KD, Deppe H. The
      ment and evaluation in vivo of a long-term delivery system for vap-        influence of surgical and restorative dental treatment prior to
      reotide, a somatostatin analogue. J Control Release 1998;52:205–213.       cardiac valve surgery on the long-term demand of dental treat-
35.   Graham PD, Brodbeck KJ, McHugh AJ. Phase inversion dynamics                ment: A prospective clinical study. Wien Med Wochenschr 2009;
      of PLGA solutions related to drug delivery. J Control Release              159:608–619.
      1999;58:233–245.                                                       41. Stein JM, Fickl S, Yekta SS, Hoischen U, Ocklenburg C, Smeets R.
36.   Allen TM, Cullis PR. Drug delivery systems: Entering the main-             Clinical evaluation of a biphasic calcium composite grafting mate-
      stream. Science 2004;303:1818–1822.                                        rial in the treatment of human periodontal intrabony defects: A
37.   Zolnik BS, Burgess DJ. Effect of acidic pH on PLGA microsphere             12-month randomized controlled clinical trial. J Periodontol 2009;
      degradation and release. J Control Release 2007;122:338–344.               80:1774–1782.

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | JAN 2011 VOL 96, ISSUE 1                                                                           185

Shared By: