978

Slow Release of TetracyclineHydrochloride From a CelluloseMembrane Used in Guided TissueRegenerationCristina Markman* Sergio E.L. Fracalanzza* Arthur B. Novaes, Jr.*and Arthur B. Novaes*

The objective of this study was to evaluate if the biologic membrane utilized forGTR can be impregnated by tetracycline- hydrochloride and if the chemotherapeuticagent, once impregnated, can be released in minimal inhibitory concentrations for aperiod compatible with clinical application. Initially, an in vitro study was done withcellulose membranes cut in pieces measuring 9 cm2. A volume of 100 containing a72,000 g/ml solution f tetracycline was dispensed onto each fragment, and dried for70 minutes at 37C. Four pieces measuring 0.5 cm2 were cut from different points ofthe 9 cm2 membrane (presumably, containing 400 g of tetracycline), placed in testtubes containing 4 ml of sterile deionized water, and agitated for 2 minutes. A standardcurve was made from known concentrations of tetracycline and compared to 10 ofthe test solutions obtained by the elution of the 0.5 cm2 fragments. The concentrationswere determined through the bioassay technique in 3 duplicate experiments. The samplesrecovered from the membrane fragments had a mean of 101 g/ml of tetracyclineliberated, demonstrating that the membrane was impregnated homogeneously by thechemotherapeutic agent. In a second phase, an in vivo study was carried out to determinethe length of time the drug was liberated from the membranes and at which concentra-tions, in the presence of an inflammatory process. Fourteen 0.5 cm2 fragments containing400 g of tetracycline were placed in 14 polypropylene chambers containing 200 ofthioglycolate medium. The chambers were implanted in the peritoneal cavities of 14mice, one chamber per animal, and left in from 1 to 14 days. They were then removedand the concentrations of tetracycline determined from 20 samples using a bioassay.The results showed that the antibiotic was released slowly from the 1st through the 12thday in decreasing concentrations that varied from 218 to 20.8 g/ml. The impregnatedcellulose membrane can probably be used in GTR acting as a membrane and as a slow-release device, liberating the chemotherapeutic agent in concentrations high enough toeliminate periodontopafhic microorganisms. / Periodontol 1995;66:978-983.Key Words: Membranes, artificial; membranes, barrier; tetracycline/therapeutic use;guided tissue regeneration.

Guided tissue regeneration (GTR) is an accepted thera-peutic modality for the treatment of furcation lesions1-2and interproximal defects.2 It can also be used for boneregeneration following tooth extractions and reconstruc-tion of alveolar ridges,3 and in association with osseoin-tegrated implants.4-5

Systemic antibiotics are usually prescribed as part of

Federal University of Rio de Janeiro, Brazil.'University of Sao Paulo at Ribeirao Preto, Brazil.

the surgical protocol to avoid the formation of microab-scesses that may occur around the third week of healing.6-7They may also be prescribed to avoid contamination ofthe membrane and consequent infection in situationswhere the membrane becomes exposed due to soft tissuerecession.8 Tetracycline hydrochloride (Te-HCl), used sys-temically or locally, is an antibiotic commonly recom-mended for various reasons: it is highly efficaciousagainst the majority of periodontopathic bacteria,9-11 it hasan acidic pH and can be used as a root demineralizing

Volume 66Number 11 MARKMAN, FRACALANZZA, NOVAES JR, NOVAES 979

agent,12 it inhibits human collagenase13 and bone rsorp-tion,14 and it is associated with bone formation.1516 Al-though a generally accepted mode of administration, theroutine use of systemic antibiotics17 may cause differentlevels of side-effects. In order to avoid the systemic useof antibiotics and possible side-effects, it would be idealif the membrane used for GTR could deliver the antibioticlocally.

The objective of this study was to evaluate if a mem-brane used in GTR* can be impregnated homogeneouslyin vitro by Te-HCl. We also determined in vivo if thedrug was liberated from the membrane and, if so, at whichconcentrations and for how long.

MATERIALS AND METHODS

Membrane Impregnation With Te-HCl: In VitroAnalysisThe sterile cellulose membrane was cut into pieces mea-suring 9 cm2 (3X3 cm). To each piece of the membrane100 of a 72,000 g/ml solution of Te-HCl was dis-pensed under sterile conditions such that the whole mem-brane was homogeneously covered by the solution. Themembranes were placed in Petri dishes and dried at 37C.The 9 cm2 pieces were then cut in smaller pieces mea-suring 0.5 cm2 and stocked in a desiccator at 10C.

Theoretically, each 0.5 cm2 piece should contain 400g of Te-HCl. To evaluate this hypothesis, four 0.5 cm2membranes taken from various parts of the original 9 cm2membrane were placed in four separate test tubes con-taining 4 ml sterile deionized water, and then agitated for2 minutes.

The concentration of Te-HCl in the solution was de-termined through the bioassay technique. The Mueller-Hinton agar medium containing Bacillus subtilis (ATCC6633) as the test bacteria was perforated with sterile cyl-inders measuring 6 mm in diameter. To each 6 mm hole10 of known concentrations of Te-HCl were dispensedin order to obtain a standard curve. A 10 sample re-moved from each of the test tubes containing the impreg-nated pieces of the membranes was also dispensed intoseparate empty holes. The experiment was always donein duplicate.

The dishes were incubated for 18 hours at 37C andthe diameter of the inhibition halos measured. The con-centration of the antibiotic in each of the membranes wascalculated from the mathematic curve obtained throughlinear regression analysis from the experimental points onthe standard curve.19 The experiment was repeated threetimes, each on a different day.

'Gengiflex, BioFill Produtos Biotecnologicus, Curitiba, PR, Brazil.

Slow Release of the Te-HCl From the CelluloseMembrane: In Vivo AnalysisMembranes measuring 0.5 cm2 impregnated with 400 g/ml of Te-HCl were placed into polypropylene chambersmeasuring 15 mm in length made from the body of 1 mlsyringes. One end of the chamber had been previouslysealed with a nitrocellulose filter with pores measuring0.22 . All the material utilized had been previously ster-ilized. The membrane was introduced into the chamberalong with 200 of thioglycolate medium. The chamberwas then sealed with another nitrocellulose filter with thesame pore size. The thioglycolate medium was used as adiluent because it also induces inflammation in the sur-rounding tissues as it diffuses through the pores in thenitrocellulose filter, mimicking the inflammation presentin the periodontal tissues of patients with periodontal dis-ease.

The chambers were introduced into the peritoneal cav-ity of 14 adult Swiss mice. This experiment was con-ducted according Federal University protocol for animalstudies. A 15-mm incision was made in the trichotomizedabdomen of the previously anesthetized animals, gainingaccess to the peritoneal cavity. One chamber was intro-duced into the peritoneal cavity of each of the 14 miceso that it adapted to the internal organs without damageto them. The incisions were sutured and the animals werekept warm until recovery from anesthesia. One animalwas sacrificed each day for the next 14 days. A new ac-cess to the peritoneal cavity was made to recover thechambers and to analyze the reaction of the tissues to theimplantation.

Immediately following retrieval of the chambers, theircontents were removed with a sterile tip adapted to a mi-cropipette from a small hole made in one of the nitrocel-lulose filters and transferred to a sterile tube. 20-1sample of each tube was diluted in thioglycolate medium1:10 to 1:80 in serial dilution, always in duplicate.

The concentration of Te-HCl was determined throughthe bioassay technique as described earlier, with the ex-ception that the standard curve was prepared with Te-HCldiluted in thioglycolate medium.

RESULTSThe membranes absorbed the antibiotic well without ap-parent alteration in their physical characteristics.

The in vitro experiment was designed to evaluate theuniformity of the impregnation of the membrane with Te-HCl. Four 0.5 cm2 pieces obtained from the original 9.0cm2 membrane were identified as Tl, T2, T3, and T4.They were placed in deionized sterile water, agitated forthe elution of the antibiotic, and the resultant solutiontested through the bioassay using Bacillus subtilis as theindicator bacteria. The experiment was repeated threetimes. As can be seen in Table 1, the mean diameter ofthe halos of inhibition for Tl, T2, T3, and T4 in the three

980 TETRACYCLINE-TREATED CELLULOSE MEMBRANE USE IN GTRJ Periodontol

November 1995

Table 1. Evolution of the Homogeneous Distribultion of Te-HCl inthe Membranes Analyzed Through the Bioassay

1000Concentrations ofTE + HCLOg/ml)

Diameter of the Halo of Inhibition (in mm)Experiment 1 Experiment 2 Experiment 3

3.12*6.25

12.5025.0050.00

100.00200.00400.00TVT2T,T4Mean T, and T,

I

>.

ta

100

10

ca

coo

1

Experiment 1Experiment 2Experiment 3

J_10 20 30 40

Size of the halo of Inhibition (mm)Figure 1. Mathematical curve obtained through linear regression anal-ysis of the experimental points in the standard curve.

ities of the chamber had adhered to the intestinal loops.The tissues in that area showed hyperemia and edema,indicating the beginning of a discrete inflammatory pro-cess. The process increased from the second to the fourthday, when a membranous capsule surrounded the chamberand the presence of a highly vascularized inflammatorytissue was present, especially close to the extremities ofthe chamber. The process remained stable up to the 14thday.

Table 2. Determination of the Concentrations of Te-HCl in the Fragments of the Impregnated Mem-branes Through the Equation Obtained Through Linear Regression Analysis of the ExperimentalPoints

Concentrations of Te-HCl in the Fragment After Elution'Experiment Equation rxy* T, T2 T3 T4 Mean1 log y = 0.21 + 0.072.x* 0.99 120.0 102.0 120,0 102.0 111.02 log y = 0.16 + 0.072.x 0.99 91.2 107.0 91.2 91.2 95.03 log y = -0.47 + 0.087.x 0.99 112.0 93.0 93.0 98.0 98.0

101.0

*Coefficient of correlation.'Each fragment was eluted in 4 ml of water.'Standard errors for Equations 1, 2 and 3 are respectively 0.06, 0.07 and 0.08. All are significant at the 1%level (P < 0,01).

\ 'In g/ml.

Volume 66Number 11 MARKMAN, FRACALANZZA, NOVAES JR, NOVAES 981

Table 3. Determination of the Concentrations of Te-HCl in theChambers Implanted in the Peritoneal Cavity Following the Indi-cated Period

Days the ChambersRemained Concentration of Te-HCl (in g/ml)

1 2182 913 1144 915 916 897 448 389 54

10 3911 2212 2113

982 TETRACYCLINE-TREATED CELLULOSE MEMBRANE USE IN GTRJ Periodontol

November 1995

released into an aqueous medium, similar to the mediumfound in the periodontal pocket.

In the in vivo phase, an experimental model with theliberation of the antibiotic, in the presence of a "sterile"inflammatory process, was successfully created. As ex-pected, the thioglycolate medium induced a local inflam-matory process, causing the formation of a capsule ofhighly vascularized newly formed tissue around and at-tached to the extremities of the chamber. The nitrocellu-lose filters at the extremities of the chamber allowed thediffusion of the antibiotic from within the chamber to thetissues around the chamber, similar to that which occurswhen a slow-release device is introduced into a periodon-tal pocket.

It is important to determine how long effective con-centrations of the antibiotic can be detected. Fourteen an-imals received one chamber each and were sacrificed dai-ly so that the concentration of the drug could be deter-mined during this period.

A decrease in the concentration of Te-HCl from 218g/ml after 24 hours to 20.8 g/ml after 12 days wasobserved. These results are highly significant due to thehigh concentration of the antibiotic that was obtained andby the way it was slowly liberated from the membrane.If used as a slow-release device in a periodontal pocketit would have been effective against the bacteria involvedin periodontal disease. The microorganisms commonlyfound in association with Periodontitis are generally sus-ceptible to a concentration of 8 g/ml of Te-HCl.9 Evenmicroorganisms such as S. sputigena, non-pigmentedBacteroides species, and some Gram-negative rods, whichhave a minimal inhibitory concentration less than or equalto 16 g/ml, could have been inhibited up to the 12thday by the impregnated cellulose membrane.

Although a comparison between the cellulose mem-brane impregnated with Te-HCl with slow-release devicesdescribed in the literature is inadequate due to differencesin the concentrations used, in the type of material em-ployed, and due to differences in the treatment protocol,the concentrations obtained and maintained for up to 12days in this study are equal to or better than the resultsobtained in some studies with slow-release devices.22 24"27

The results obtained with the methodology employedin this study indicate that concentrations between 218 and20 g/ml up to 12 days allow adequate control of thebacteria involved in the disease and the microorganismsthat might adhere to the membrane in case of membraneexposure during the healing period.ConclusionsThe methodology employed for the impregnation of thecellulose membrane was adequate, reproducible, and easyto perform, important factors to make serial productionfeasible.

The cellulose membrane impregnated with tetracycline

can be utilized in periodontal procedures, because in ad-dition to being a biocompatible membrane, it acts as abarrier during GTR and has the capability of liberatingTe-HCl up to 12 days in concentrations more than suffi-cient to eliminate periodontopathic microorganisms.

REFERENCES1. Gottlow J, Nyman S, Karring T, Lindhe J. New attachment forma-

tion as the result of controlled tissue regeneration. J Clin Periodon-tol 1984;11:494-503.

2. Becker W, Becker BE, Prichard JE, Caffesse E, Rosenberg E. Newattachment with root isolation procedures: Report of Class III andClass II furcation and vertical osseous defects. Int J PeriodonticsRestorative Dent 1988;8:9-23.

3. Nevins M, Mellonig JT. The advantages of localized ridge augmen-tation prior to implant placement: A staged event. Int J PeriodonticsRestorative Dent 1994;14:97-111.

4. Novaes Jr. AB, Novaes AB. IMZ implants placed into extractionsockets in association with membrane therapy (Gengiflex) and po-rous hydroxyapatite: A case report. Int J Oral Maxillofac Implants1992;7:536-540.

5. Novaes Jr. AB, Novaes AB. Bone formation over an IMZ implantplaced into an extraction socket in association with membrane ther-apy (Gengiflex). Clin Oral Implant Res 1993;4:106-110.

6. Dahlin C, Lekholm U, Linde A. Membrane-induced bone augmen-tation at titanium implants. A report on ten fixtures followed from1 to 3 years after loading. Ini J Periodontics Restorative Dent1991;11:273-281.

7. Becker W, Becker BE, Handelsman M, Ochsenbein C, AlbrektssonT. Guided tissue regeneration for implants placed into extractionsockets: A study in dogs. J Periodontol 1991;62:703-709.

8. Demolon IA, Persson GR, Monda BJ, Johnson RH, Ammons WEEffects of antibiotic treatment on clinical conditions and bacterialgrowth with guided tissue regeneration. J Periodontol 1993;64:609-616.

9. Genco RJ. Antibiotics in the treatment of human periodontal dis-eases. J Periodontol 1981;52:545-558.

10. Walker CB, Gordon JM, Socransky S. Antibiotic susceptibility test-ing of subgingival plaque samples. J Clin Periodontol 1983;10:422-432.

11. Walker CB, Pappas JD, Tyler KZ, Cohen S, Gordon JM. Approachesto chemotherapy. III. Antibiotic susceptibilities of periodontal bac-teria. "In vitro" susceptibilities to eight antimicrobial agents. J Per-iodontol 1985;56:67-74.

12. Wikesj UME, Christersson LA, Genco RJ, et al. A biochemicalapproach to periodontal regeneration: Tetracycline treatment condi-tions dentin surfaces. J Periodont Res 1986;21:322-329.

13. Golub LM, Ramamurthy N, McNamara TF, et al. Tetracyclines in-hibit tissue collagenase activity. A new mechanism in the treatmentof periodontal disease. J Periodont Res 1984;19:651-655.

14. Rifkin BR, Vernilo AT, Golub LM. Blocking periodontal diseaseprogression by inhibiting tissue-destructive enzymes: A potentialtherapeutic role for tetracyclines and their chemically-modified an-alogs. J Periodontol 1993;64:819-827.

15. Sasaki T, Ramamurthy NS, Golub LM. Tetracycline administrationincreases collagen synthesis in osteoblasts of diabetic rats: A quan-titative autoradiographic study. Calcif Tissue Int 1992,50:411^119.

16. Mabry TW, Yukna RA, Sepe WW. Freeze-dried bone allograftscombined with tetracycline in the treatment of juvenile Periodontitis.J Periodontol 1985;56:74-81.

17. McGhee JR, Michalek SM, Cassel GH. Dental Microbiology. Phil-adelphia: Harper & Row Publishers, Inc.; 1982:155.

18. Novaes Jr. AB, Novaes AB, Grisi MFM, Soares UN, Gabarra F.

Volume 66Number 11 MARKMAN, FRACALANZZA, NOVAES JR, NOVAES 983

Gengiflex. An alkali-cellulose membrane for GTR. Histologie ob-servations. Braz Dent J 1993;4:65-71.

19. Ipsen J, Feigl P. Bancroft's Introduction to Biostatistics. New York:Harper & Row Publishers; 95-102, 2nd ed.

20. Ory EM. The tetraeyclines. In: Kagan BM, ed. Antimicrobial Ther-apy, 3rd ed. Philadelphia: W.B. Saunders Co.; 1980:117-126.

21. Silverstein L, Bissada N, Manoucher-Pour M, Greenwell H. Clinicaland microbiologie effects of tetraeycline irrigation on Periodontitis.J Periodontol 1988;59:301-305.

22. Goodson JM, Haffajee A, Socransky SS. Periodontal therapy bylocal delivery of tetraeycline. J Clin Periodontol 1979;6:83-92.

23. Lindhe J, Heijl L, Goodson JM, Socransky SS. Local tetraeyclinedelivery using hollow fiber devices in periodontal therapy. J ClinPeriodontol 1979;6:141-149.

24. Goodson JM, Offenbacher S, Fair DH, Hogan PE. Periodontal dis-

ease treatment by local drug delivery. J Periodontol 1985;56:265-272.

25. Goodson JM, Holborow D, Dunn RL, Hogan PE, Dunham S. Mon-olithic tetracycline-containing fibers for controlled delivery to peri-odontal pockets. J Periodontol 1983;54:575-579.

26. Baker PJ, Evans RT, Slots J, Genco RJ. Susceptibility of human oralanaerobic bacteria to antibiotic suitable for topical use. J Clin Per-iodontol 1981;12:201-208.

27. Walker CB, Gordon JM, McQuilkin SJ, Niebloom TA, SocranskySS. Tetracycline: Levels achievable in gingival crevice fluid and "invitro" effect on subgingival organisms. Part II. Susceptibilities ofperiodontal bacteria. J Periodontol 1981;52:613-616.

Send reprint requests to: Dr. Arthur B. Novaes, Jr., Av. das Americas1155, Room 1002, Rio de Janeiro, RJ, Brazil 22631-000.

Accepted for publication May 11, 1995.