- Email: [email protected]
polyglycolide copolymer in bone defect healing in humans
Available online at www.sciencedirect.com
Biomaterials 29 (2008) 1817e1823 www.elsevier.com/locate/biomaterials
Polylactide/polyglycolide copolymer in bone defect healing in humans Carlo Bertoldi a, Davide Zaffe b,*, Ugo Consolo a a
Department of Neurosciences, HeadeNeck and Rehabilitation, Section of Dentistry and Maxillofacial Surgery, University of Modena and Reggio Emilia, Modena, Italy b Department of Anatomy and Histology, Section of Human Anatomy, University of Modena and Reggio Emilia, Via del Pozzo, 71-41100 Modena, Italy Received 26 September 2007; accepted 25 December 2007 Available online 29 January 2008
Abstract This pilot study aims to evaluate the healing of a large defects in the human jawbone filled with a Poly-Lactide-co-Glycolide (PLG) polymer (FisiograftÒ) by means of clinical, radiological and histological methods and to compare the results with those of platelet-rich plasma (PRP) clot or autologous bone (AB) fillings. Bone cysts, where previous non-surgical treatments failed to promote healing, underwent surgery. Nineteen consenting male patients were randomly split into three groups, packed with PRP, AB or PLG. A core biopsy was performed 4 and 6 months after surgery. All treated defects showed clinical, radiological and histological progresses over time. AB provided the best clinical and histological performance and PLG had overlapping outcomes; PRP filling was statistically different. Six months after surgery, bone activities were enhanced in sites treated with PLG and fairly good with PRP. Additionally, PLG showed some new lamellar formations. In conclusion, outcomes were best with AB graft, but suitable results were achieved using PLG to promote healing of severe bone defects. PLG shows only a delayed regenerative capability but does not require a secondary donor site. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: Autologous bone; Bone regeneration; Copolymer; In vivo test; Polyglycolic acid; Polylactic acid
1. Introduction Bone regeneration of large bony defects or fractures remains even today a substantial therapeutic challenge. Grafting by means of biomaterials is often needed to achieve bone healing. Autologous bone is currently the gold standard graft material for reconstruction procedures [1], but some disadvantages, such as the need for a second surgical site, limits its use. Synthetic biomaterials show clinical promise for the treatment of intra-osseous defects [2,3] without the risk of immunogenic and infectious complications. Particular attention has been focused on polymeric biomaterials because of their easy handling, good biocompatibility, and interesting biomechanical properties [4,5]. Some polymeric biomaterials are resorbable and are active on bone processes, thus allowing coherent
* Corresponding author. Tel.: þ39 0594224800; fax: þ39 0594224861. E-mail address: [email protected] (D. Zaffe). 0142-9612/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2007.12.034
bone regeneration without the need to supplemental interventions to remove the implanted material [3,6]. The families of polylactic acid (PLA), polyglycolyc acid (PGA) and their co-polymers (Poly-Lactide-co-Glycolide, PLG), both biodegradable, are at the cutting edge of bone reconstruction procedures. These materials are manufactured in a variety of forms to produce appliances for use in bone surgery such as tissue barriers [7], fixation devices [8,9], porous solid graft [3,5,10e12], solid or semi-solid carriers for delivery of growth factor (GF), bone morphogenetic proteins (BMPs) and other bioactive molecules [13,14]. These materials may also serve as scaffolding to promote specific cell adhesion, differentiation and bone formation [5,12,14,15]. This polymer is gradually substituted by new bone over time, but its degradation depends on the formulation, amorphous/crystalline structure, isomeric characteristics, molecular weight and amount of material used [10,16]. The exact composition and architecture of the polymeric scaffold becomes fundamental to promote successful bone regeneration.
1818
C. Bertoldi et al. / Biomaterials 29 (2008) 1817e1823
Although extensively used in several orthopedic appliances [17], PLG have been scarcely applied in oro-maxillo-facial applications. PLA-derived devices were implemented to prevent alveolar osteitis or dry socket with discordant results by Hooley and Golden [18]. PLA and decalcified freeze-dried bone allografts were clinically studied in the treatment of periodontal intra-osseous defects by Meadows et al. [19]. Results showed a limited amount of tissue regeneration and the persistence of PLA particles surrounded by soft tissue. A new PLG (FisiograftÒ, Ghimas, Italy), manufactured as sponge blocks, gel and powders, was studied in maxillary sinus floor augmentations [12,20], as space maker, in alveolar ridge preservation [5], ridge augmentation with split-crest technique, in the treatment of deep periodontal intra-osseous defect [11], and in critical bone defects in experimental animals [3] with contrasting results. Little data on the use of PLG in bone defects to promote spontaneous bone healing in the human jaw is currently available [15]. The aim of this pilot study was the clinical, radiological and histological comparison of large human jawbone defects that fail to heal spontaneously in the short to medium term, treated with PLG, autologous bone (AB) or platelet-rich plasma (PRP) clot. 2. Materials and methods Patients were recruited from referrals to outpatient clinic for large intrabone jaw defects secondary to periapical radicular cysts [21] at the Department of Neurosciences, HeadeNeck and Rehabilitation, University of Modena and Reggio Emilia. The local District Ethical Committee approved the study. Only non-smokers were included. Subjects younger than 18 years and/or with a history of severe acute or chronic systemic or topical disease, pregnant or lactating women and subjects taking medications known to affect platelets, oral and bone status were excluded. All patients presented at least one periapical radicular cyst in relation to teeth (or, more frequently, to residue roots of teeth) with a hopeless prognosis for subsequent prosthetics or when previous non-surgical treatments failed to achieve healing [22]. The pre-treatment and radiological diagnosis of periapical radicular cyst were not considered sufficient [23] for enrollment if, after surgery, the periapical radicular cyst was not confirmed by histopathological examination [21,22]. Cystic lesions of different origin or etiopathogenesis were excluded. Intra-osseous lesion had a diameter ranging from 1 to 3 cm. Patients were randomly assigned to three different groups using a random number generator software (Random Allocation Software, M. Saghaei, Isfahan, Iran). Cysts were filled with: CG1 (control group 1) e platelet-rich plasma (PRP) clot, prepared by the Immuno-transfusion Department of the Modena Medical Center [24]; CG2 (control group 2) e autogenous bone grafting; TG (test group) e PLG (FisiograftÒ, Ghimas s.r.l., Casalecchio di Reno BO, Italy), 50% DL lactic acid and 50% glycolic acid (50 PLA : 50 PGA) mixed with dextran as excipient. Blood (about 450 ml), drawn 24 h prior to surgery in a blood container containing 80 ml acid citrate dextrose solution, was centrifuged (RC3C, Sorvall, U.S.A.) at 1200g for 6 min at 20 C to obtain fraction 1 (about 280 ml), plasma and platelets, and fraction 2 (about 250 ml), erythrocytes. Immediately prior to surgery, fraction 1 was centrifuged at 4400g for 6 min at 14 C to obtain platelet-poor plasma (PPP, 240 ml) and platelet-rich plasma (PRP, 40 5 ml of) fractions. Part of the PRP fraction was used to prepare autologous thrombin by adding calcium chloride (recalcification). PRP activation [24] and gelation was obtained by mixing PRP with autologous thrombin (3:1).
Autologous bone was harvested from the anterior iliac crest [24]. Conserving the inguinal ligament and femoral cutaneous nerve, a linear incision of soft tissue, along the outer border of the iliac crest (1 cm behind the anterior superior iliac spine), was performed. PLG was used as sponge block, gel and powders, following indications by the manufacturer [25] and suggestions in Katanec et al. [15], to improve handling and promote cyst cavity filling. In all residual defects, a FisiograftÒ sponge block of adequate size was initially positioned in the cavity. The sponge block was surrounded by FisiograftÒ gel mixed with powder (50:50) to fill the residual space [15,25]. Blinded clinical and radiological assessments were performed by conventional radiographs (e.g. orthopantomography) and CT with orthogonal reconstruction implemented by specific software (dental scan). Before surgery, the CT radiologic section showing the greatest size of the defect was chosen in each patient. The radiologist matched the examination performed prior the trial with the same during follow-up. Cysts were surgically approached using standardized technique [26]. Filler was applied to entirely pack the bone defect without compression. Patients underwent regular clinical and radiological assessment to track the progress of the treatment for up to 6 months. On radiographs, the area, maximum and minimum diameters and mean density level of each defect was measured using a suitable image analyzer and software (AnalySISÒ, Soft Imaging System GmbH, Mu¨nster, Germany). Additionally, the mean density of a fixed cortical-bone area of uniform gray aspect outside the defect (the same site for each patient) was evaluated in radiographs of each stage to compute the site of treatment/bone (S/B) density rate. Each parameter was expressed as the ratio R between post-treatment/pre-treatment phases for each patient. Comparisons were performed by means of non-parametric ManneWhitney U test, KruskaleWallis followed by the StudenteNewmaneKeuls multiple comparison test [27]. The null hypothesis H0 was rejected for a critical significance level of P < 0.05. Cylindrical core biopsies were obtained with a hollow mill at 600g under saline jet after 4 and 6 months. Biopsies were fixed in 4% paraformaldehyde (all reagents came from Fluka Chemie AG, Switzerland) in 0.1 M phosphate buffer pH 7.2 for 2 h at room temperature. Specimens [12,28] were dehydrated through ethanol series at 4 C, then embedded in methyl methacrylate (PMMA) using a water bath at 4 C. The PMMA blocks were serially sectioned along the longitudinal axis of the biopsy up to its center using a diamond saw microtome (1600, Leica, Germany). A set of 5 mm-thick sections was obtained starting from this level using a tungsten carbide knife (Profile D) on a bone microtome (Autocut 1150, Reichert-Jung, Austria). Then a thick section (200 mm) was obtained from the center of each biopsy using the diamond saw microtome. Thin sections were singly stained with toluidine blue, trichrome Gomori stain, solochrome cyanine/congo red, alkaline phosphatase (ALP) histochemical method, a marker of osteogenesis (azo dye ¼ Fast Blue BB), or tartrate resistant acid phosphatase (TRAP) histochemical method, a marker of osteoclasts (azo dye ¼ Fast Red RC). Thick sections, which were reduced by grinding to 100 mm and perfectly polished with emery paper and alumina, were X-ray microradiographed (Italstructures, Italy) at 8 kV and 12.5 mA on Kodak SO 343 film. The microradiographs and the sections were analyzed and photographed using an Axiophot microscope (Carl Zeiss AG, Oberkochen, Germany) under ordinary and polarized light. Trabecular bone volume (TBV) of biopsies, an index of the amount of bone tissue [12,28], was evaluated on the microradiographs using a suitable program for image analyzer and software (AnalySISÒ, Soft Imaging System GmbH, Mu¨nster, Germany).
3. Results Nineteen men, ranging in age from 24 to 67 years (mean S.D. ¼ 39.6 12.1), completed the trial (Table 1). Two patients, randomly inserted in the TG (test group, packed with PLG), presented with two bilateral cysts. These patients underwent two biopsies: the first 4 months and the second 6 months after surgery (Table 1). Four treated lesions were biopsied one after 4 months and three after 6 months in each group
C. Bertoldi et al. / Biomaterials 29 (2008) 1817e1823
1819
Table 1 Age (years), cyst size (mm2), groups and biopsy time (months) of the 19 patients Patient
Age
Size
Group
Biopsy
A.C. C.B. C.V. D.B. D.F. D.G. E.O. F.C. F.F. F.P.
41 26 25 44 42 34 32 42 67 39
G.F. G.M. L.C. L.L. L.R. M.F. M.P. P.V.
38 56 29 63 33 39 24 47
R.P.
31
95 104.5 109.3 181.8 75.2 134.05 161.2 187.7 116.1 102.1 157.2 81.5 159.5 178.9 90.6 161.8 132.8 146.5 183.3 107.4 64
CG2 CG2 CG1 CG2 CG2 CG2 CG2 CG1 CG2 TG TG CG1 TG TG CG1 CG1 CG1 CG1 TG TG TG
6 6 6 4 4 6 4 6 4 4 6 4 4 6 4 6 4 4 4 6 4
(Table 1). Patient age was not statistically different (P ¼ 0.74, KruskaleWallis test) across the three groups: CG1 (37.7 13.1); CG2 (40.8 13.4), TG (40.4 11.3). No treated patients had complications due to surgery. The PRP produced by the patient’s whole blood (starting concentration of 300,000 50,000 platelets/ml) contained 1,000,000 250,000 platelets/ml; therefore platelet concentration increased more than threefold. The size and radiopacity variations before and 4 months after surgery in the three groups are shown in Fig. 1. The most relevant feature is enlargement of the defect in CG1. After 4 months, the gray level (0e255) was higher in TG and CG2, with some topical differences. Fig. 1 illustrates the behavior of the index ratios R (ratio of the parameter value after surgery to that before surgery) of size and gray level of treated defects. Four months after treatment, CG1 showed greater (R > 1) size parameters (Area, Dmax, Dmin) and lower (R ¥ 1) gray level values than those of TG and CG2. After 6 months, CG1 had reduced size (R < 1) and increased gray level (R > 1), even if in lower amounts than those recorded in the other groups. CG2, after both 4 and 6 months, had lower size values than TG, with the exception of Dmin after 4 months, whereas gray level was greater than TG only after 6 months. Consequently, TG had size parameters slightly greater than those of CG2 (excluding the Dmin) but always lower than those of CG1 after 4 months. At this time, the TG gray level was the greatest. After 6 months, TG size parameters were always greater and gray levels were lower than those of CG2. Non-parametric statistical tests pointed out a statistical significance between CG1 parameters versus both CG2 and TG parameters only after 4 months. Longitudinal (4 versus 6 months) non-parametric analysis detected for size a statistical difference in CG1 and CG2, whereas gray level was only statistically significant in CG1.
Fig. 1. Graph showing the behavior of four parameters (area (A), maximum (D) and minimum (d) diameter, and gray level ratio S/B, see Section 2) of cyst defects of CG1, CG2, and TG, 4 and 6 months after surgery. The S/B corresponds to the mean ratio between the gray level of the cyst defect and of a fixed cortical-bone area of uniform aspect in the radiograph of each patient. Each column corresponds to the mean ratio (R) between each parameter value after surgery to the corresponding parameter value before surgery. Note the size parameter reduction and the gray level increase with time across all groups. Four months after surgery, CG2 and TG are always statistically (*P < 0.05) lower (size) and greater (gray level) than CG1. Six-months size parameters are statistically ( P < 0.05) lower than those at 4 months in CG1 and CG2. Six-month density rate was statistically ( P < 0.05) greater than that at 4 months in CG1 only.
Four months after surgery, histology highlighted the nearly complete absence of bone formation in CG1. Biopsies of this group were almost completely formed by dense fibrous tissue (Fig. 2a). Some osteoclasts, eroding minute remnants of bone, were observed inside the dense fibrous tissue (Fig. 2b). A lone bone trabecula (Fig. 2c) most likely originating from the bone wall was found inside one of the biopsies. The trabecula showed an indented surface without osteoclasts and feeble histochemical ALP expression (Fig. 2d) in some cells (osteoblasts) laid down on its surface. CG2 biopsies showed the contemporary presence of autologous bone fragments surrounded by fibrous tissue (Fig. 2e) and new bone formed in apposition to autologous bone fragments or, more frequently, newly formed by static osteogenesis [29]. Autologous bone fragments completely surrounded by fibrous tissue had trabeculae with roughly eroded surface of and some osteoclasts forming Howship’s lacunae (Fig. 2f). Biopsies of TG showed bone, newly formed by static osteogenesis [29], growing among PLG particles (Fig. 2g). Fibroblasts surrounded the PLG particles as epithelial-like cells (Fig. 2g and h). These epithelial-like cells were distinguishable from bone-lining cells only when a space persisted between bone and particles (Fig. 2h). Osteogenetic activity was frequently found in the form of osteoblast presence (Fig. 2i) and/or ALP expression (Fig. 2j). Six months after surgery, histology highlighted a small presence of bone inside CG1 (Fig. 3a). All bone had a nearly woven
1820
C. Bertoldi et al. / Biomaterials 29 (2008) 1817e1823
Fig. 2. Histology of CG1 (aed), CG2 (eef) and TG (gej) biopsies taken 4 months after surgery. Toluidine blue (c, e, i), trichrome Gomori (a, g), solochrome cyanine/congo red (h) staining, TRAP (b, f) and ALP (d, j) methods. Note in (a) the highly fibrous appearance of the CG1 biopsy. The green arrows in (b) point to osteoclasts resorbing small bone remnants. The yellow arrows in (c) and blue arrows in (d) point to disordered osteoblasts, ALP positive, adhering to the lone bone trabecula of a CG1 biopsy. The yellow arrow in (e) points to a bone trabecula newly formed by static osteogenesis, and green arrows in (f) point to osteoclasts resorbing an autologous bone fragment of CG2. Note in (g) the newly formed bone in between PLG (FisiograftÒ) particles (F) of TG. In (h) two cell layers, one lining the bone and the other surrounding the PLG particle, are shown. In (i) and (j) several osteoblasts and ALP expression of the newly formed bone (B) of TG are shown. Field width ¼ 420 mm (a, b, g); 720 mm (c, d); 850 mm (e, f); 270 mm (h); 310 mm (i, j).
fibered structure (Fig. 3b), presented a randomized ALP expression and small osteoclastic activities, providing for feeble remodeling (Fig. 3c). A great amount of bone was observed in biopsies of CG2 (Fig. 3d). Some large autologous fragments (Fig. 3e) spared by osteoclasts (Fig. 3f) in displayed lamellar structure with a few islands of woven-structured newly formed bone (Fig. 3e). An appreciable amount of bone was observed in biopsies of TG (Fig. 3g). Almost all trabeculae had woven structure (Fig. 3h); inter-trabecular fibrous tissue presented a strong ALP expression (Fig. 3i) whereas osteoclasts were scanty. PLG particles were highly reduced, but some particles could still be recognized 6 months after surgery (Fig. 3j). A small amount of lamellar bone was newly deposited in apposition to previously formed woven bone (Fig. 3j). Quantitative evaluation of the amount of mineralized tissue (Fig. 4) by computerized measurements on microradiograph of the central section of each biopsy (expressed as trabecular bone volume e TBV), revealed a significant difference between the values of the filled defects after 4 (mean: CG1 ¼ 0%; CG2 ¼ 16.5%; TG ¼ 16.2) and 6 (mean: CG1 ¼ 8.0%; CG2 ¼ 36.0%; TG ¼ 30.1) months across all groups (P < 0.04, ManneWhitney U test). Four months after surgery, the KruskaleWallis test highlighted a statistical significance (P ¼ 0.022) among the three groups. The post-hoc Studente NewmaneKeuls multiple comparison tests detected that both the CG1/CG2 and CG1/TG contrasts were statistically significant (P < 0.05), but no statistical difference was found for CG2
and TG. Contrary to the evaluations at 4 months, no statistical significance (KruskaleWallis test) was found for TBV values of all groups 6 months after surgery. 4. Discussion Healing of bone defects depends on the osteogenic potential of host bone, but the biological species, anatomical zone, and size of defect are often limiting factors. Bone regeneration is particularly arduous in severe bone defects. A periapical bone cysts can be considered a ‘‘critical defect’’ [2,14] when its initial mean diameter exceeds 1 cm [15]. If not grafted, these defects do not spontaneously heal 6 months after surgery [30]. The bone cysts we selected for grafting were inflammatory defects that surrounded the apex of teeth or more frequently residual roots of teeth having a hopeless prognosis or defects that failed to regenerate after endodontic clinical therapy alone and could only be managed by a careful surgical approach [15,21,26]. Defect grafting was an option to improve bone healing since surgical treatment alone would require more than 1 year before definitive results were achieved [30]. Due to ethical reasons, the higher regeneration potentialities of PRP in tissue healing [24] demanded the use of PRP instead of blood clot in control group 1 to assure a certain degree of bone repair within 4e6 months. The PRP produced by the patient’s blood contained a platelet concentration of about 3.4fold. Marx [31] and Consolo et al. [28] used the same platelet
C. Bertoldi et al. / Biomaterials 29 (2008) 1817e1823
1821
Fig. 3. Histology of CG1 (aec), CG2 (def) and TG (gej) biopsies taken 6 months after surgery. Microradiograph (a, d, g), Toluidine blue (h, j), TRAP, ordinary light (c, f) and polarized light (b, e), and ALP (i) methods. Note in (a) the scanty new bone in CG1 biopsy. This bone has a mainly woven structure (b) with few osteoclasts (c e green arrows). Note in (d) the appreciable amount of bone in CG2 biopsy. The yellow arrow in (e) points to the only part of woven bone, newly formed in apposition to autologous bone, spared from osteoclast (f e green arrows) erosion. Note in (g) the good amount of bone of a TG biopsy. This bone (B) has nearly a woven structure, osteoblasts (h) and the surrounding fibrous tissue show good ALP expression (i). The red arrow in (j) points to an osteoblast line, between two PLG (FisiograftÒ) particles (F), which form lamellar bone (LB) in apposition to previously formed woven bone (WB e yellow arrow). Field width ¼ 2300 mm (a); 1050 mm (b, c); 4650 mm (d, g); 790 mm (e, f, h, i); 675 mm (j).
concentration, preparation and activation methods and achieved good results. Moreover, this platelet concentration seems to guarantee better effects on cultured fibroblast and osteoblast [32] than higher or lower concentrations. This control group was partnered with the CG2 (autologous bone) to verify the osteoregenerative properties of the two natural grafts: the former with low and the latter, the gold standard in dentistry [1], with optimal capabilities. The choice of expressing each parameter (area, maximum and minimum diameters, gray level) as the ratio of the change in values after and before the surgery was made to normalize the results of evaluations across all patients and to refer the results to the initial state. Radiographic evaluation indicated a substantial difference between CG1 and the other two groups 4 months after surgery. Moreover, we recorded a statistically significant improvement of bone regeneration in this group 6 months after surgery. The great increase in fibrous tissue was probably the main drawback to the bone regeneration process, as shown by osteoclast erosion of the scanty bone formation. The improvement observed after 6 months may be ascribable to PRP, which might have a feeble but positive effect on cells of the osteogenic lineage [24]. As expected, the CG2 (autologous bone) group had the best results, but TG outcomes overlapped those of CG2 after both 4 and 6 months. Statistical significance was achieved for size parameter of CG2, both after 4 and 6 months. No statistical
significance was found longitudinally comparing gray level values of CG2 and all values of TG. Histology highlights the good features of TG, after both 4 and 6 months, but with lower osteogenic potential as compared with that of CG2 after 6 months. Bone, mostly due to static osteogenesis (bone characterized by cords of pluristratified stationary osteoblasts, each polarized in a different direction), i.e. the former process of intramembranous ossification [29], is irregularly formed in between PLG particles. Bone formation in TG occurs with a disordered and scattered trabecular displacement, also observed with the use of PLG for sinus floor augmentation in man [12]. Notwithstanding, in TG new bone regularly had a woven structure and a few parts had a lamellar structure, due to dynamic osteogenesis (bone formed by monostratified laminae of movable osteoblasts, all polarized in the same direction [29]), indicating a relatively fast speed of bone formation. PLG had both appreciable osteoregenerative capabilities and the absence of inflammatory response. This is probably due to its composition, which is not a PLA (polylactic acid) polymer but a Poly-Lactic-co-Glycolic polymer. Each polymer has different behaviors and biological responses dependent on its manufacturing process (crystallization degree, reticulation level, etc.) and not only on its chemical composition [10,19,33]. PLA polymers produce inflammatory effects [19,33] and have longer degradation times, sometimes over 5 years [33]. PLG are chemically and biologically different
1822
C. Bertoldi et al. / Biomaterials 29 (2008) 1817e1823
handling of the material, associated with adequate new bone formation. Larger clinical and radiographic studies are warranted to exactly define the time required by PLG to complete bone defects healing. Acknowledgements The authors wish to thank Dr. John Pradelli, MD for assistance in manuscript draft and revision. The Research University Fund (FAR) of the Department of Neurosciences, HeadeNeck and Rehabilitation, University of Modena and Reggio Emilia supported this investigation. References Fig. 4. Graph showing the TBV amount (expressed as bone percentage (%) of the whole biopsy) in CG1 (PRP e -), CG2 (AO e :) and TG (PLG e C) biopsies, 4 and 6 months after surgery. The superimposed lines reveal the TBV time trend.
from PLA polymers, but their features can be extremely variable depending on chemical composition and polymer preparation [10]. The PLG we tested (FisiograftÒ) is a 50e50 copolymer whose preparation yields a material that is free of inflammation effects, seems to stimulate bone growth, and decomposes after 6e8 months [20]. Even if some authors ascribe only a mechanical effect to PLG [16] our results showed a strong ALP expression, indicating a probable promotion effect on bone cells, since similar results are not achieved using a material which has mechanical effects alone (e.g. zirconia particles). Our results confirm the behavior of PLG observed in experimental animals and in non-critical size defects in humans. Hollinger [34] observed good bone repair in the short term after PLG grafting in rats. Rimondini et al. [3] found a continual bone formation after FisiograftÒ grafting in critical defects of rabbits for up to 90 days. In humans, Minenna et al. [11] did not find additional benefits in terms of clinical attachment level gain and probing depth reduction with the addition of FisiografÒ for treating deep intra-osseous defects with open flap debridement procedures. Serino et al. [5] in sockets, Zaffe et al. [12] and Scarano et al. [20] in sinus floor augmentations reported satisfactory results using FisiograftÒ. FisiograftÒ grafting of periradicular cyst defects was clinically, radiographically and densitometrically studied by Katanec et al. [15], who reported satisfactory results in 84% of patients. The results of our study, though admitting the superiority of autologous bone, support the suitability of PLG grafting, which is associated with satisfactory bone formation though requiring slightly longer times than autologous bone. In conclusion, our work extends the validity of PLG grafts to the treatment of human bone defects, as already assessed in critical defects in animals. The advantage of PLG with respect to the autologous bone lies in the absence of donor-site lesions, decreased post-operative morbidity, shortened surgical times, absence of cross infections and availability and ease of
[1] Zijderveld SA, Zerbo IR, van den Bergh JP, Schulten EA, ten Bruggenkate CM. Maxillary sinus floor augmentation using a betatricalcium phosphate (Cerasorb) alone compared to autogenous bone grafts. Int J Oral Maxillofac Implants 2005;20:432e40. [2] Trombelli L, Heitz-Mayfield LJA, Needleman I, Moles D, Scabbia A. A systematic review of graft materials and biological agents for periodontal intraosseous defects. J Clin Periodontol 2002;29(Suppl. 3):117e35. [3] Rimondini L, Nicoli-Aldini N, Fini M, Guzzardella G, Tschon M, Giardino R. In vivo experimental study on bone regeneration in critical bone defects using an injectable biodegradable PLA/PGA copolymer. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2005; 99:148e54. [4] Griffith LG. Polymeric biomaterials. Acta Mater 2000;48:263e77. [5] Serino G, Biancu S, Iezzi G, Piattelli A. Ridge preservation following tooth extraction using a polylactide and polyglycolide sponge as space filler: a clinical and histological study in humans. Clin Oral Implants Res 2003;14:651e8. [6] Trisi P, Rao W, Rebaudi A, Fiore P. Histologic effect of pure-phase betatricalcium phosphate on bone regeneration on human artificial jaw bone defects. Int J Periodontics Restorative Dent 2003;23:69e77. [7] Hammerle CH, Lang NP. Single stage surgery combining transmucosal implant placement with guided bone regeneration and bioresorbable materials. Clin Oral Implants Res 2001;12:9e18. [8] Peltoniemi H, Ashammakhi N, Kontio R, Waris T, Salo A, Linqvist C, et al. The use of bioadsorbable osteofixation devices in craniomaxillofacial surgery. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2002;94:5e14. [9] Leiggener CS, Curtis R, Mu¨ller AA, Pfluger D, Gogolewski S, Rahn BA. Influence of copolymer composition of polylactide implants on cranial bone regeneration. Biomaterials 2006;27:202e7. [10] Martin C, Winet H, Bao JH. Acidity near eroding polylactideepolyglycolide in vitro and in vivo rabbit tibial bone chambers. Biomaterials 1996;17:2373e80. [11] Minenna L, Herrero F, Sanz M, Trombelli L. Adjunctive effect of a polylactide/polyglycolide copolymer in the treatment of deep periodontal intra-osseous defects: a randomized clinical trial. J Clin Periodontol 2005;32:456e61. [12] Zaffe D, Leghissa GC, Pradelli J, Botticelli AR. Histological study on sinus lift grafting by FisiograftÒ and Bio-Oss. J Mater Sci Mater Med 2005;16:789e93. [13] El-Amin SF, Lu HH, Kahn Y, Burems J, Mithchell J, Tuan RS, et al. Extracellular matrix production by human osteoblasts cultured on biodegradable polymers applicable for tissue engineering. Biomaterials 2003;24:1213e21. [14] Saito N, Okada T, Horiuchi H, Ota H, Takahashi J, Murakami N, et al. Local bone formation by injection of recombinant human bone morphogenetic protein-2 contained in polymer carrier. Bone 2003;32:381e6. [15] Katanec D, Pavelic´ B, Ivasovic´ Z. Efficiency of polylactide/polyglycolide copolymers bone replacement in bone defects healing measured by densitometry. Coll Antropol 2004;1:331e6.
C. Bertoldi et al. / Biomaterials 29 (2008) 1817e1823 [16] Lu HH, El-Amin SF, Scott KD, Laurencin CT. Three-dimensional, bioactive, biodegradable, polymer-bioactive glass composite scaffold with improved mechanical properties support collagen synthesis and mineralization of human osteoblast-like cells in vitro. J Biomed Mater Res 2003;64(A):465e74. [17] Athanasiou KA, Agrawal CM, Barber F, Burkart SS. Orthopedic applications for PLAePGA biodegradable polymers. Arthroscopy 1998;14:726e37. [18] Hooley JR, Golden DP. The effect of polylactic acid granule on the incidence of alveolar osteitis after mandibular third molar surgery. Oral Surg Oral Medi Oral Pathol 1995;80:279e83. [19] Meadows CL, Gher ME, Quintero G, Lafferty TA. A comparison of polylactic acid granules and decalcified freeze-dried bone allograft in human periodontal osseous defects. J Periodontol 1993;64:103e9. [20] Scarano A, Degidi M, Iezzi G, Pecora G, Piattelli M, Orsini G, et al. Maxillary sinus augmentation with different biomaterials: a comparative histologic and histomorphometric study in man. Implant Dent 2006; 15:197e205. [21] Nair PNR. On the causes of persistent apical periodontitis: a review. Int Endod J 2006;39:249e81. [22] Nair PNR. New prospectives on radicular cysts: do they heal. Int Endod J 1998;31:155e60. [23] Shrout MK, Hall JM, Hildebolt CE. Differentiation of periapical granulomas and radicular cysts by digital radiometric analysis. Oral Surg Oral Medi Oral Pathol 1993;76:356e61. [24] Consolo U, Zaffe D, Bertoldi C, Ceccherelli G. PRP activity on maxillary sinus floor augmentation by autologous bone. Clin Oral Implant Res 2007;18:252e62. [25] Appendix to Certificate of Assessment N. 2001-OSL-MDD-0069 (Fisiograft, CE0434, Class III devices), Det Norske Veritas Region Norge, Høvik (Norway), 20 February 2001.
1823
[26] Walton RE. Endodontic surgery ‘‘Principles of endodontics’’. Chapter. 18. In: Peterson LJ, Ellis E, Hupp JR, editors. Contemporary oral and maxillofacial surgery. 3rd ed.). St. Louis: Mosby; 1998. p. 433e55. [27] Glantz SA, editor. Primer of biostatistics. 5th ed. New York: Mc-Graw Hill; 2003. p. 1e487. [28] Consolo U, Bertoldi C, Zaffe D. Intermittent loading improves results in mandibular alveolar distraction osteogenesis. Clin Oral Implant Res 2006;17:179e87. [29] Ferretti M, Palumbo C, Contri M, Marotti G. Static and dinamic osteogenesis: two different types of bone formation. Anat Embryol 2002;206:21e9. [30] Mitchell R. An evaluation of bone healing in cavities in the jaws implanted with a collagen matrix. Brit J Oral Maxillofac Surg 1992;30:180e2. [31] Marx RE. Platelet-rich plasma: a source of multiple autologous growth factors for bone grafts. In: Lynch SE, Genco RJ, Marx RE, editors. Tissue engineering: application in maxillofacial surgery and periodontics. Chicago: Quintessence Publishing Co.; 1999. p. 71e82. [32] Graziani F, Ivanovski S, Cei S, Ducci F, Tonetti M, Gabriele M. The in vitro effect of different PRP concentrations on osteoblasts and fibroblasts. Clin Oral Implants Res 2006;17:212e9. [33] Kontio R, Ruuttila P, Lindroos L, Suuronen R, Salo A, Lindqvist C, et al. Biodegradable polydioxanone and poly(l/d)lactide implants: an experimental study on peri-implant tissue response. Int J Oral Maxillofac Surg 2005;34:766e76. [34] Hollinger JO. Preliminary report on the osteogenic potential of a biodegradable copolymer of polylactide (PLA) and polyglycolide (PGA). J Biomed Mater Res 1983;17:71e82.
Biomaterials 29 (2008) 1817e1823 www.elsevier.com/locate/biomaterials
Polylactide/polyglycolide copolymer in bone defect healing in humans Carlo Bertoldi a, Davide Zaffe b,*, Ugo Consolo a a
Department of Neurosciences, HeadeNeck and Rehabilitation, Section of Dentistry and Maxillofacial Surgery, University of Modena and Reggio Emilia, Modena, Italy b Department of Anatomy and Histology, Section of Human Anatomy, University of Modena and Reggio Emilia, Via del Pozzo, 71-41100 Modena, Italy Received 26 September 2007; accepted 25 December 2007 Available online 29 January 2008
Abstract This pilot study aims to evaluate the healing of a large defects in the human jawbone filled with a Poly-Lactide-co-Glycolide (PLG) polymer (FisiograftÒ) by means of clinical, radiological and histological methods and to compare the results with those of platelet-rich plasma (PRP) clot or autologous bone (AB) fillings. Bone cysts, where previous non-surgical treatments failed to promote healing, underwent surgery. Nineteen consenting male patients were randomly split into three groups, packed with PRP, AB or PLG. A core biopsy was performed 4 and 6 months after surgery. All treated defects showed clinical, radiological and histological progresses over time. AB provided the best clinical and histological performance and PLG had overlapping outcomes; PRP filling was statistically different. Six months after surgery, bone activities were enhanced in sites treated with PLG and fairly good with PRP. Additionally, PLG showed some new lamellar formations. In conclusion, outcomes were best with AB graft, but suitable results were achieved using PLG to promote healing of severe bone defects. PLG shows only a delayed regenerative capability but does not require a secondary donor site. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: Autologous bone; Bone regeneration; Copolymer; In vivo test; Polyglycolic acid; Polylactic acid
1. Introduction Bone regeneration of large bony defects or fractures remains even today a substantial therapeutic challenge. Grafting by means of biomaterials is often needed to achieve bone healing. Autologous bone is currently the gold standard graft material for reconstruction procedures [1], but some disadvantages, such as the need for a second surgical site, limits its use. Synthetic biomaterials show clinical promise for the treatment of intra-osseous defects [2,3] without the risk of immunogenic and infectious complications. Particular attention has been focused on polymeric biomaterials because of their easy handling, good biocompatibility, and interesting biomechanical properties [4,5]. Some polymeric biomaterials are resorbable and are active on bone processes, thus allowing coherent
* Corresponding author. Tel.: þ39 0594224800; fax: þ39 0594224861. E-mail address: [email protected] (D. Zaffe). 0142-9612/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2007.12.034
bone regeneration without the need to supplemental interventions to remove the implanted material [3,6]. The families of polylactic acid (PLA), polyglycolyc acid (PGA) and their co-polymers (Poly-Lactide-co-Glycolide, PLG), both biodegradable, are at the cutting edge of bone reconstruction procedures. These materials are manufactured in a variety of forms to produce appliances for use in bone surgery such as tissue barriers [7], fixation devices [8,9], porous solid graft [3,5,10e12], solid or semi-solid carriers for delivery of growth factor (GF), bone morphogenetic proteins (BMPs) and other bioactive molecules [13,14]. These materials may also serve as scaffolding to promote specific cell adhesion, differentiation and bone formation [5,12,14,15]. This polymer is gradually substituted by new bone over time, but its degradation depends on the formulation, amorphous/crystalline structure, isomeric characteristics, molecular weight and amount of material used [10,16]. The exact composition and architecture of the polymeric scaffold becomes fundamental to promote successful bone regeneration.
1818
C. Bertoldi et al. / Biomaterials 29 (2008) 1817e1823
Although extensively used in several orthopedic appliances [17], PLG have been scarcely applied in oro-maxillo-facial applications. PLA-derived devices were implemented to prevent alveolar osteitis or dry socket with discordant results by Hooley and Golden [18]. PLA and decalcified freeze-dried bone allografts were clinically studied in the treatment of periodontal intra-osseous defects by Meadows et al. [19]. Results showed a limited amount of tissue regeneration and the persistence of PLA particles surrounded by soft tissue. A new PLG (FisiograftÒ, Ghimas, Italy), manufactured as sponge blocks, gel and powders, was studied in maxillary sinus floor augmentations [12,20], as space maker, in alveolar ridge preservation [5], ridge augmentation with split-crest technique, in the treatment of deep periodontal intra-osseous defect [11], and in critical bone defects in experimental animals [3] with contrasting results. Little data on the use of PLG in bone defects to promote spontaneous bone healing in the human jaw is currently available [15]. The aim of this pilot study was the clinical, radiological and histological comparison of large human jawbone defects that fail to heal spontaneously in the short to medium term, treated with PLG, autologous bone (AB) or platelet-rich plasma (PRP) clot. 2. Materials and methods Patients were recruited from referrals to outpatient clinic for large intrabone jaw defects secondary to periapical radicular cysts [21] at the Department of Neurosciences, HeadeNeck and Rehabilitation, University of Modena and Reggio Emilia. The local District Ethical Committee approved the study. Only non-smokers were included. Subjects younger than 18 years and/or with a history of severe acute or chronic systemic or topical disease, pregnant or lactating women and subjects taking medications known to affect platelets, oral and bone status were excluded. All patients presented at least one periapical radicular cyst in relation to teeth (or, more frequently, to residue roots of teeth) with a hopeless prognosis for subsequent prosthetics or when previous non-surgical treatments failed to achieve healing [22]. The pre-treatment and radiological diagnosis of periapical radicular cyst were not considered sufficient [23] for enrollment if, after surgery, the periapical radicular cyst was not confirmed by histopathological examination [21,22]. Cystic lesions of different origin or etiopathogenesis were excluded. Intra-osseous lesion had a diameter ranging from 1 to 3 cm. Patients were randomly assigned to three different groups using a random number generator software (Random Allocation Software, M. Saghaei, Isfahan, Iran). Cysts were filled with: CG1 (control group 1) e platelet-rich plasma (PRP) clot, prepared by the Immuno-transfusion Department of the Modena Medical Center [24]; CG2 (control group 2) e autogenous bone grafting; TG (test group) e PLG (FisiograftÒ, Ghimas s.r.l., Casalecchio di Reno BO, Italy), 50% DL lactic acid and 50% glycolic acid (50 PLA : 50 PGA) mixed with dextran as excipient. Blood (about 450 ml), drawn 24 h prior to surgery in a blood container containing 80 ml acid citrate dextrose solution, was centrifuged (RC3C, Sorvall, U.S.A.) at 1200g for 6 min at 20 C to obtain fraction 1 (about 280 ml), plasma and platelets, and fraction 2 (about 250 ml), erythrocytes. Immediately prior to surgery, fraction 1 was centrifuged at 4400g for 6 min at 14 C to obtain platelet-poor plasma (PPP, 240 ml) and platelet-rich plasma (PRP, 40 5 ml of) fractions. Part of the PRP fraction was used to prepare autologous thrombin by adding calcium chloride (recalcification). PRP activation [24] and gelation was obtained by mixing PRP with autologous thrombin (3:1).
Autologous bone was harvested from the anterior iliac crest [24]. Conserving the inguinal ligament and femoral cutaneous nerve, a linear incision of soft tissue, along the outer border of the iliac crest (1 cm behind the anterior superior iliac spine), was performed. PLG was used as sponge block, gel and powders, following indications by the manufacturer [25] and suggestions in Katanec et al. [15], to improve handling and promote cyst cavity filling. In all residual defects, a FisiograftÒ sponge block of adequate size was initially positioned in the cavity. The sponge block was surrounded by FisiograftÒ gel mixed with powder (50:50) to fill the residual space [15,25]. Blinded clinical and radiological assessments were performed by conventional radiographs (e.g. orthopantomography) and CT with orthogonal reconstruction implemented by specific software (dental scan). Before surgery, the CT radiologic section showing the greatest size of the defect was chosen in each patient. The radiologist matched the examination performed prior the trial with the same during follow-up. Cysts were surgically approached using standardized technique [26]. Filler was applied to entirely pack the bone defect without compression. Patients underwent regular clinical and radiological assessment to track the progress of the treatment for up to 6 months. On radiographs, the area, maximum and minimum diameters and mean density level of each defect was measured using a suitable image analyzer and software (AnalySISÒ, Soft Imaging System GmbH, Mu¨nster, Germany). Additionally, the mean density of a fixed cortical-bone area of uniform gray aspect outside the defect (the same site for each patient) was evaluated in radiographs of each stage to compute the site of treatment/bone (S/B) density rate. Each parameter was expressed as the ratio R between post-treatment/pre-treatment phases for each patient. Comparisons were performed by means of non-parametric ManneWhitney U test, KruskaleWallis followed by the StudenteNewmaneKeuls multiple comparison test [27]. The null hypothesis H0 was rejected for a critical significance level of P < 0.05. Cylindrical core biopsies were obtained with a hollow mill at 600g under saline jet after 4 and 6 months. Biopsies were fixed in 4% paraformaldehyde (all reagents came from Fluka Chemie AG, Switzerland) in 0.1 M phosphate buffer pH 7.2 for 2 h at room temperature. Specimens [12,28] were dehydrated through ethanol series at 4 C, then embedded in methyl methacrylate (PMMA) using a water bath at 4 C. The PMMA blocks were serially sectioned along the longitudinal axis of the biopsy up to its center using a diamond saw microtome (1600, Leica, Germany). A set of 5 mm-thick sections was obtained starting from this level using a tungsten carbide knife (Profile D) on a bone microtome (Autocut 1150, Reichert-Jung, Austria). Then a thick section (200 mm) was obtained from the center of each biopsy using the diamond saw microtome. Thin sections were singly stained with toluidine blue, trichrome Gomori stain, solochrome cyanine/congo red, alkaline phosphatase (ALP) histochemical method, a marker of osteogenesis (azo dye ¼ Fast Blue BB), or tartrate resistant acid phosphatase (TRAP) histochemical method, a marker of osteoclasts (azo dye ¼ Fast Red RC). Thick sections, which were reduced by grinding to 100 mm and perfectly polished with emery paper and alumina, were X-ray microradiographed (Italstructures, Italy) at 8 kV and 12.5 mA on Kodak SO 343 film. The microradiographs and the sections were analyzed and photographed using an Axiophot microscope (Carl Zeiss AG, Oberkochen, Germany) under ordinary and polarized light. Trabecular bone volume (TBV) of biopsies, an index of the amount of bone tissue [12,28], was evaluated on the microradiographs using a suitable program for image analyzer and software (AnalySISÒ, Soft Imaging System GmbH, Mu¨nster, Germany).
3. Results Nineteen men, ranging in age from 24 to 67 years (mean S.D. ¼ 39.6 12.1), completed the trial (Table 1). Two patients, randomly inserted in the TG (test group, packed with PLG), presented with two bilateral cysts. These patients underwent two biopsies: the first 4 months and the second 6 months after surgery (Table 1). Four treated lesions were biopsied one after 4 months and three after 6 months in each group
C. Bertoldi et al. / Biomaterials 29 (2008) 1817e1823
1819
Table 1 Age (years), cyst size (mm2), groups and biopsy time (months) of the 19 patients Patient
Age
Size
Group
Biopsy
A.C. C.B. C.V. D.B. D.F. D.G. E.O. F.C. F.F. F.P.
41 26 25 44 42 34 32 42 67 39
G.F. G.M. L.C. L.L. L.R. M.F. M.P. P.V.
38 56 29 63 33 39 24 47
R.P.
31
95 104.5 109.3 181.8 75.2 134.05 161.2 187.7 116.1 102.1 157.2 81.5 159.5 178.9 90.6 161.8 132.8 146.5 183.3 107.4 64
CG2 CG2 CG1 CG2 CG2 CG2 CG2 CG1 CG2 TG TG CG1 TG TG CG1 CG1 CG1 CG1 TG TG TG
6 6 6 4 4 6 4 6 4 4 6 4 4 6 4 6 4 4 4 6 4
(Table 1). Patient age was not statistically different (P ¼ 0.74, KruskaleWallis test) across the three groups: CG1 (37.7 13.1); CG2 (40.8 13.4), TG (40.4 11.3). No treated patients had complications due to surgery. The PRP produced by the patient’s whole blood (starting concentration of 300,000 50,000 platelets/ml) contained 1,000,000 250,000 platelets/ml; therefore platelet concentration increased more than threefold. The size and radiopacity variations before and 4 months after surgery in the three groups are shown in Fig. 1. The most relevant feature is enlargement of the defect in CG1. After 4 months, the gray level (0e255) was higher in TG and CG2, with some topical differences. Fig. 1 illustrates the behavior of the index ratios R (ratio of the parameter value after surgery to that before surgery) of size and gray level of treated defects. Four months after treatment, CG1 showed greater (R > 1) size parameters (Area, Dmax, Dmin) and lower (R ¥ 1) gray level values than those of TG and CG2. After 6 months, CG1 had reduced size (R < 1) and increased gray level (R > 1), even if in lower amounts than those recorded in the other groups. CG2, after both 4 and 6 months, had lower size values than TG, with the exception of Dmin after 4 months, whereas gray level was greater than TG only after 6 months. Consequently, TG had size parameters slightly greater than those of CG2 (excluding the Dmin) but always lower than those of CG1 after 4 months. At this time, the TG gray level was the greatest. After 6 months, TG size parameters were always greater and gray levels were lower than those of CG2. Non-parametric statistical tests pointed out a statistical significance between CG1 parameters versus both CG2 and TG parameters only after 4 months. Longitudinal (4 versus 6 months) non-parametric analysis detected for size a statistical difference in CG1 and CG2, whereas gray level was only statistically significant in CG1.
Fig. 1. Graph showing the behavior of four parameters (area (A), maximum (D) and minimum (d) diameter, and gray level ratio S/B, see Section 2) of cyst defects of CG1, CG2, and TG, 4 and 6 months after surgery. The S/B corresponds to the mean ratio between the gray level of the cyst defect and of a fixed cortical-bone area of uniform aspect in the radiograph of each patient. Each column corresponds to the mean ratio (R) between each parameter value after surgery to the corresponding parameter value before surgery. Note the size parameter reduction and the gray level increase with time across all groups. Four months after surgery, CG2 and TG are always statistically (*P < 0.05) lower (size) and greater (gray level) than CG1. Six-months size parameters are statistically ( P < 0.05) lower than those at 4 months in CG1 and CG2. Six-month density rate was statistically ( P < 0.05) greater than that at 4 months in CG1 only.
Four months after surgery, histology highlighted the nearly complete absence of bone formation in CG1. Biopsies of this group were almost completely formed by dense fibrous tissue (Fig. 2a). Some osteoclasts, eroding minute remnants of bone, were observed inside the dense fibrous tissue (Fig. 2b). A lone bone trabecula (Fig. 2c) most likely originating from the bone wall was found inside one of the biopsies. The trabecula showed an indented surface without osteoclasts and feeble histochemical ALP expression (Fig. 2d) in some cells (osteoblasts) laid down on its surface. CG2 biopsies showed the contemporary presence of autologous bone fragments surrounded by fibrous tissue (Fig. 2e) and new bone formed in apposition to autologous bone fragments or, more frequently, newly formed by static osteogenesis [29]. Autologous bone fragments completely surrounded by fibrous tissue had trabeculae with roughly eroded surface of and some osteoclasts forming Howship’s lacunae (Fig. 2f). Biopsies of TG showed bone, newly formed by static osteogenesis [29], growing among PLG particles (Fig. 2g). Fibroblasts surrounded the PLG particles as epithelial-like cells (Fig. 2g and h). These epithelial-like cells were distinguishable from bone-lining cells only when a space persisted between bone and particles (Fig. 2h). Osteogenetic activity was frequently found in the form of osteoblast presence (Fig. 2i) and/or ALP expression (Fig. 2j). Six months after surgery, histology highlighted a small presence of bone inside CG1 (Fig. 3a). All bone had a nearly woven
1820
C. Bertoldi et al. / Biomaterials 29 (2008) 1817e1823
Fig. 2. Histology of CG1 (aed), CG2 (eef) and TG (gej) biopsies taken 4 months after surgery. Toluidine blue (c, e, i), trichrome Gomori (a, g), solochrome cyanine/congo red (h) staining, TRAP (b, f) and ALP (d, j) methods. Note in (a) the highly fibrous appearance of the CG1 biopsy. The green arrows in (b) point to osteoclasts resorbing small bone remnants. The yellow arrows in (c) and blue arrows in (d) point to disordered osteoblasts, ALP positive, adhering to the lone bone trabecula of a CG1 biopsy. The yellow arrow in (e) points to a bone trabecula newly formed by static osteogenesis, and green arrows in (f) point to osteoclasts resorbing an autologous bone fragment of CG2. Note in (g) the newly formed bone in between PLG (FisiograftÒ) particles (F) of TG. In (h) two cell layers, one lining the bone and the other surrounding the PLG particle, are shown. In (i) and (j) several osteoblasts and ALP expression of the newly formed bone (B) of TG are shown. Field width ¼ 420 mm (a, b, g); 720 mm (c, d); 850 mm (e, f); 270 mm (h); 310 mm (i, j).
fibered structure (Fig. 3b), presented a randomized ALP expression and small osteoclastic activities, providing for feeble remodeling (Fig. 3c). A great amount of bone was observed in biopsies of CG2 (Fig. 3d). Some large autologous fragments (Fig. 3e) spared by osteoclasts (Fig. 3f) in displayed lamellar structure with a few islands of woven-structured newly formed bone (Fig. 3e). An appreciable amount of bone was observed in biopsies of TG (Fig. 3g). Almost all trabeculae had woven structure (Fig. 3h); inter-trabecular fibrous tissue presented a strong ALP expression (Fig. 3i) whereas osteoclasts were scanty. PLG particles were highly reduced, but some particles could still be recognized 6 months after surgery (Fig. 3j). A small amount of lamellar bone was newly deposited in apposition to previously formed woven bone (Fig. 3j). Quantitative evaluation of the amount of mineralized tissue (Fig. 4) by computerized measurements on microradiograph of the central section of each biopsy (expressed as trabecular bone volume e TBV), revealed a significant difference between the values of the filled defects after 4 (mean: CG1 ¼ 0%; CG2 ¼ 16.5%; TG ¼ 16.2) and 6 (mean: CG1 ¼ 8.0%; CG2 ¼ 36.0%; TG ¼ 30.1) months across all groups (P < 0.04, ManneWhitney U test). Four months after surgery, the KruskaleWallis test highlighted a statistical significance (P ¼ 0.022) among the three groups. The post-hoc Studente NewmaneKeuls multiple comparison tests detected that both the CG1/CG2 and CG1/TG contrasts were statistically significant (P < 0.05), but no statistical difference was found for CG2
and TG. Contrary to the evaluations at 4 months, no statistical significance (KruskaleWallis test) was found for TBV values of all groups 6 months after surgery. 4. Discussion Healing of bone defects depends on the osteogenic potential of host bone, but the biological species, anatomical zone, and size of defect are often limiting factors. Bone regeneration is particularly arduous in severe bone defects. A periapical bone cysts can be considered a ‘‘critical defect’’ [2,14] when its initial mean diameter exceeds 1 cm [15]. If not grafted, these defects do not spontaneously heal 6 months after surgery [30]. The bone cysts we selected for grafting were inflammatory defects that surrounded the apex of teeth or more frequently residual roots of teeth having a hopeless prognosis or defects that failed to regenerate after endodontic clinical therapy alone and could only be managed by a careful surgical approach [15,21,26]. Defect grafting was an option to improve bone healing since surgical treatment alone would require more than 1 year before definitive results were achieved [30]. Due to ethical reasons, the higher regeneration potentialities of PRP in tissue healing [24] demanded the use of PRP instead of blood clot in control group 1 to assure a certain degree of bone repair within 4e6 months. The PRP produced by the patient’s blood contained a platelet concentration of about 3.4fold. Marx [31] and Consolo et al. [28] used the same platelet
C. Bertoldi et al. / Biomaterials 29 (2008) 1817e1823
1821
Fig. 3. Histology of CG1 (aec), CG2 (def) and TG (gej) biopsies taken 6 months after surgery. Microradiograph (a, d, g), Toluidine blue (h, j), TRAP, ordinary light (c, f) and polarized light (b, e), and ALP (i) methods. Note in (a) the scanty new bone in CG1 biopsy. This bone has a mainly woven structure (b) with few osteoclasts (c e green arrows). Note in (d) the appreciable amount of bone in CG2 biopsy. The yellow arrow in (e) points to the only part of woven bone, newly formed in apposition to autologous bone, spared from osteoclast (f e green arrows) erosion. Note in (g) the good amount of bone of a TG biopsy. This bone (B) has nearly a woven structure, osteoblasts (h) and the surrounding fibrous tissue show good ALP expression (i). The red arrow in (j) points to an osteoblast line, between two PLG (FisiograftÒ) particles (F), which form lamellar bone (LB) in apposition to previously formed woven bone (WB e yellow arrow). Field width ¼ 2300 mm (a); 1050 mm (b, c); 4650 mm (d, g); 790 mm (e, f, h, i); 675 mm (j).
concentration, preparation and activation methods and achieved good results. Moreover, this platelet concentration seems to guarantee better effects on cultured fibroblast and osteoblast [32] than higher or lower concentrations. This control group was partnered with the CG2 (autologous bone) to verify the osteoregenerative properties of the two natural grafts: the former with low and the latter, the gold standard in dentistry [1], with optimal capabilities. The choice of expressing each parameter (area, maximum and minimum diameters, gray level) as the ratio of the change in values after and before the surgery was made to normalize the results of evaluations across all patients and to refer the results to the initial state. Radiographic evaluation indicated a substantial difference between CG1 and the other two groups 4 months after surgery. Moreover, we recorded a statistically significant improvement of bone regeneration in this group 6 months after surgery. The great increase in fibrous tissue was probably the main drawback to the bone regeneration process, as shown by osteoclast erosion of the scanty bone formation. The improvement observed after 6 months may be ascribable to PRP, which might have a feeble but positive effect on cells of the osteogenic lineage [24]. As expected, the CG2 (autologous bone) group had the best results, but TG outcomes overlapped those of CG2 after both 4 and 6 months. Statistical significance was achieved for size parameter of CG2, both after 4 and 6 months. No statistical
significance was found longitudinally comparing gray level values of CG2 and all values of TG. Histology highlights the good features of TG, after both 4 and 6 months, but with lower osteogenic potential as compared with that of CG2 after 6 months. Bone, mostly due to static osteogenesis (bone characterized by cords of pluristratified stationary osteoblasts, each polarized in a different direction), i.e. the former process of intramembranous ossification [29], is irregularly formed in between PLG particles. Bone formation in TG occurs with a disordered and scattered trabecular displacement, also observed with the use of PLG for sinus floor augmentation in man [12]. Notwithstanding, in TG new bone regularly had a woven structure and a few parts had a lamellar structure, due to dynamic osteogenesis (bone formed by monostratified laminae of movable osteoblasts, all polarized in the same direction [29]), indicating a relatively fast speed of bone formation. PLG had both appreciable osteoregenerative capabilities and the absence of inflammatory response. This is probably due to its composition, which is not a PLA (polylactic acid) polymer but a Poly-Lactic-co-Glycolic polymer. Each polymer has different behaviors and biological responses dependent on its manufacturing process (crystallization degree, reticulation level, etc.) and not only on its chemical composition [10,19,33]. PLA polymers produce inflammatory effects [19,33] and have longer degradation times, sometimes over 5 years [33]. PLG are chemically and biologically different
1822
C. Bertoldi et al. / Biomaterials 29 (2008) 1817e1823
handling of the material, associated with adequate new bone formation. Larger clinical and radiographic studies are warranted to exactly define the time required by PLG to complete bone defects healing. Acknowledgements The authors wish to thank Dr. John Pradelli, MD for assistance in manuscript draft and revision. The Research University Fund (FAR) of the Department of Neurosciences, HeadeNeck and Rehabilitation, University of Modena and Reggio Emilia supported this investigation. References Fig. 4. Graph showing the TBV amount (expressed as bone percentage (%) of the whole biopsy) in CG1 (PRP e -), CG2 (AO e :) and TG (PLG e C) biopsies, 4 and 6 months after surgery. The superimposed lines reveal the TBV time trend.
from PLA polymers, but their features can be extremely variable depending on chemical composition and polymer preparation [10]. The PLG we tested (FisiograftÒ) is a 50e50 copolymer whose preparation yields a material that is free of inflammation effects, seems to stimulate bone growth, and decomposes after 6e8 months [20]. Even if some authors ascribe only a mechanical effect to PLG [16] our results showed a strong ALP expression, indicating a probable promotion effect on bone cells, since similar results are not achieved using a material which has mechanical effects alone (e.g. zirconia particles). Our results confirm the behavior of PLG observed in experimental animals and in non-critical size defects in humans. Hollinger [34] observed good bone repair in the short term after PLG grafting in rats. Rimondini et al. [3] found a continual bone formation after FisiograftÒ grafting in critical defects of rabbits for up to 90 days. In humans, Minenna et al. [11] did not find additional benefits in terms of clinical attachment level gain and probing depth reduction with the addition of FisiografÒ for treating deep intra-osseous defects with open flap debridement procedures. Serino et al. [5] in sockets, Zaffe et al. [12] and Scarano et al. [20] in sinus floor augmentations reported satisfactory results using FisiograftÒ. FisiograftÒ grafting of periradicular cyst defects was clinically, radiographically and densitometrically studied by Katanec et al. [15], who reported satisfactory results in 84% of patients. The results of our study, though admitting the superiority of autologous bone, support the suitability of PLG grafting, which is associated with satisfactory bone formation though requiring slightly longer times than autologous bone. In conclusion, our work extends the validity of PLG grafts to the treatment of human bone defects, as already assessed in critical defects in animals. The advantage of PLG with respect to the autologous bone lies in the absence of donor-site lesions, decreased post-operative morbidity, shortened surgical times, absence of cross infections and availability and ease of
[1] Zijderveld SA, Zerbo IR, van den Bergh JP, Schulten EA, ten Bruggenkate CM. Maxillary sinus floor augmentation using a betatricalcium phosphate (Cerasorb) alone compared to autogenous bone grafts. Int J Oral Maxillofac Implants 2005;20:432e40. [2] Trombelli L, Heitz-Mayfield LJA, Needleman I, Moles D, Scabbia A. A systematic review of graft materials and biological agents for periodontal intraosseous defects. J Clin Periodontol 2002;29(Suppl. 3):117e35. [3] Rimondini L, Nicoli-Aldini N, Fini M, Guzzardella G, Tschon M, Giardino R. In vivo experimental study on bone regeneration in critical bone defects using an injectable biodegradable PLA/PGA copolymer. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2005; 99:148e54. [4] Griffith LG. Polymeric biomaterials. Acta Mater 2000;48:263e77. [5] Serino G, Biancu S, Iezzi G, Piattelli A. Ridge preservation following tooth extraction using a polylactide and polyglycolide sponge as space filler: a clinical and histological study in humans. Clin Oral Implants Res 2003;14:651e8. [6] Trisi P, Rao W, Rebaudi A, Fiore P. Histologic effect of pure-phase betatricalcium phosphate on bone regeneration on human artificial jaw bone defects. Int J Periodontics Restorative Dent 2003;23:69e77. [7] Hammerle CH, Lang NP. Single stage surgery combining transmucosal implant placement with guided bone regeneration and bioresorbable materials. Clin Oral Implants Res 2001;12:9e18. [8] Peltoniemi H, Ashammakhi N, Kontio R, Waris T, Salo A, Linqvist C, et al. The use of bioadsorbable osteofixation devices in craniomaxillofacial surgery. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2002;94:5e14. [9] Leiggener CS, Curtis R, Mu¨ller AA, Pfluger D, Gogolewski S, Rahn BA. Influence of copolymer composition of polylactide implants on cranial bone regeneration. Biomaterials 2006;27:202e7. [10] Martin C, Winet H, Bao JH. Acidity near eroding polylactideepolyglycolide in vitro and in vivo rabbit tibial bone chambers. Biomaterials 1996;17:2373e80. [11] Minenna L, Herrero F, Sanz M, Trombelli L. Adjunctive effect of a polylactide/polyglycolide copolymer in the treatment of deep periodontal intra-osseous defects: a randomized clinical trial. J Clin Periodontol 2005;32:456e61. [12] Zaffe D, Leghissa GC, Pradelli J, Botticelli AR. Histological study on sinus lift grafting by FisiograftÒ and Bio-Oss. J Mater Sci Mater Med 2005;16:789e93. [13] El-Amin SF, Lu HH, Kahn Y, Burems J, Mithchell J, Tuan RS, et al. Extracellular matrix production by human osteoblasts cultured on biodegradable polymers applicable for tissue engineering. Biomaterials 2003;24:1213e21. [14] Saito N, Okada T, Horiuchi H, Ota H, Takahashi J, Murakami N, et al. Local bone formation by injection of recombinant human bone morphogenetic protein-2 contained in polymer carrier. Bone 2003;32:381e6. [15] Katanec D, Pavelic´ B, Ivasovic´ Z. Efficiency of polylactide/polyglycolide copolymers bone replacement in bone defects healing measured by densitometry. Coll Antropol 2004;1:331e6.
C. Bertoldi et al. / Biomaterials 29 (2008) 1817e1823 [16] Lu HH, El-Amin SF, Scott KD, Laurencin CT. Three-dimensional, bioactive, biodegradable, polymer-bioactive glass composite scaffold with improved mechanical properties support collagen synthesis and mineralization of human osteoblast-like cells in vitro. J Biomed Mater Res 2003;64(A):465e74. [17] Athanasiou KA, Agrawal CM, Barber F, Burkart SS. Orthopedic applications for PLAePGA biodegradable polymers. Arthroscopy 1998;14:726e37. [18] Hooley JR, Golden DP. The effect of polylactic acid granule on the incidence of alveolar osteitis after mandibular third molar surgery. Oral Surg Oral Medi Oral Pathol 1995;80:279e83. [19] Meadows CL, Gher ME, Quintero G, Lafferty TA. A comparison of polylactic acid granules and decalcified freeze-dried bone allograft in human periodontal osseous defects. J Periodontol 1993;64:103e9. [20] Scarano A, Degidi M, Iezzi G, Pecora G, Piattelli M, Orsini G, et al. Maxillary sinus augmentation with different biomaterials: a comparative histologic and histomorphometric study in man. Implant Dent 2006; 15:197e205. [21] Nair PNR. On the causes of persistent apical periodontitis: a review. Int Endod J 2006;39:249e81. [22] Nair PNR. New prospectives on radicular cysts: do they heal. Int Endod J 1998;31:155e60. [23] Shrout MK, Hall JM, Hildebolt CE. Differentiation of periapical granulomas and radicular cysts by digital radiometric analysis. Oral Surg Oral Medi Oral Pathol 1993;76:356e61. [24] Consolo U, Zaffe D, Bertoldi C, Ceccherelli G. PRP activity on maxillary sinus floor augmentation by autologous bone. Clin Oral Implant Res 2007;18:252e62. [25] Appendix to Certificate of Assessment N. 2001-OSL-MDD-0069 (Fisiograft, CE0434, Class III devices), Det Norske Veritas Region Norge, Høvik (Norway), 20 February 2001.
1823
[26] Walton RE. Endodontic surgery ‘‘Principles of endodontics’’. Chapter. 18. In: Peterson LJ, Ellis E, Hupp JR, editors. Contemporary oral and maxillofacial surgery. 3rd ed.). St. Louis: Mosby; 1998. p. 433e55. [27] Glantz SA, editor. Primer of biostatistics. 5th ed. New York: Mc-Graw Hill; 2003. p. 1e487. [28] Consolo U, Bertoldi C, Zaffe D. Intermittent loading improves results in mandibular alveolar distraction osteogenesis. Clin Oral Implant Res 2006;17:179e87. [29] Ferretti M, Palumbo C, Contri M, Marotti G. Static and dinamic osteogenesis: two different types of bone formation. Anat Embryol 2002;206:21e9. [30] Mitchell R. An evaluation of bone healing in cavities in the jaws implanted with a collagen matrix. Brit J Oral Maxillofac Surg 1992;30:180e2. [31] Marx RE. Platelet-rich plasma: a source of multiple autologous growth factors for bone grafts. In: Lynch SE, Genco RJ, Marx RE, editors. Tissue engineering: application in maxillofacial surgery and periodontics. Chicago: Quintessence Publishing Co.; 1999. p. 71e82. [32] Graziani F, Ivanovski S, Cei S, Ducci F, Tonetti M, Gabriele M. The in vitro effect of different PRP concentrations on osteoblasts and fibroblasts. Clin Oral Implants Res 2006;17:212e9. [33] Kontio R, Ruuttila P, Lindroos L, Suuronen R, Salo A, Lindqvist C, et al. Biodegradable polydioxanone and poly(l/d)lactide implants: an experimental study on peri-implant tissue response. Int J Oral Maxillofac Surg 2005;34:766e76. [34] Hollinger JO. Preliminary report on the osteogenic potential of a biodegradable copolymer of polylactide (PLA) and polyglycolide (PGA). J Biomed Mater Res 1983;17:71e82.