SCIENTIFIC ARTICLE

In vitro antimicrobialand cytotoxic effects of Kri 1 pasteand -eugenolused in primarytooth pulpectomies Kelly J. Wright, DMDSergio V. Barbosa, DDS, PhD Kouji Araki, DDS,PhD Larz S.W. Sp~ngberg,DDS, PhD Abstract The antimicrobial and cytotoxic effects of Kri I paste, an iodoform-basedprimary tooth filling material, were comparedwith zinc oxide- (ZOE), using in vitro techniques. Antimicrobial evaluation involved measuring inhibition zones Streptococcus faecalis on brain heart agar. Cytotoxicity evaluation involved direct cell-medicamentcontact experiments of 4-hr and 24-hr duration using fresh and set medicaments,and indirect cell-medicamentcontact experiments of 24-hr duration using fresh and set medicaments.ZOE produced a greater zone of bacterial inhibition than Kri 1 paste. Kri 1 paste cytotoxicity remainedhigh regardless of the amountof setting time in the 4-hr direct contact experiment, while ZOEcytotoxicity decreased with setting time. Both Kri I paste and ZOEhad high cytotoxicity regardless of setting time in the 24-hr direct cell-medicament contact test. ZOEcytotoxicity decreased to control levels after only 1 day of setting in the indirect contact experiments, comparedwith greater than 7 days for Kri I paste. The results suggest ZOEhas better antimicrobial activity than Kri I paste. ZOEalso has lower cytotoxicity, although prolongedcell-medicament contact mayresult in both medicamentshaving similarly high cytotoxicity. (Pediatr Dent 16:102-6, 1994)

Introduction time of the material. No osteolytic changes were found To maintain function, esthetics, arch length, and arch surrounding ZOEimplants. Several studies have in- vestigated the antimicrobial action of Kri I in vivo 9’ 11-13 symmetry, primary teeth should be maintained in the 1~-16 dental arch until their proper exfoliation timeo1, 2 and in vitro. No studies have comparedthe antibac- Endodontic treatment of primary teeth has been a suc- terial action and attendant cytotoxicity of Kri 1 paste cessful method of maintaining nonvital primary teeth and ZOEin vitro. with and without periapical pathosis. However, suc- The purpose of this study was to examine the claim cess is related to strict tooth selection criteria and to of superior antimicrobial action for Kri I paste by com- thorough debridement of the canals followed by paring the antimicrobial and cytotoxic effects of Kri 1 obturation with a suitable filling material. 2, 3 paste and ZOEin vitro. Zinc oxide-eugenol (ZOE) has traditionally been used Methods and materials as a root canal filling material in the primary dentition. Recently, an iodoform paste (Kri I paste, Pharmachemie Antimicrobialevaluation AG, Zurich, Switzerland) has been recommended as an Kri 1 paste (iodoform 80.8%, camphor 4.86%, p- alternative medicament to ZOEin root canal therapy chlorophenol 2.025%, menthol 1.215%, was used as for nonvital primary teeth. ~ Walkhoff originally de- supplied. ZOE(zinc oxide U.S.P./eugenol U.S.P. -- scribed the formula for an iodoform paste in 1928 and Sultan Chemists, Inc., Englewood, NJ) was mixed to used7 it in permanentteeth. Later, Castagnola and Orlay, the same consistency as Kri I paste immediately before Juge, 8and Barker and Lockett 9reported using experimentation (2.5 g zinc-oxide to 1 ml eugenol). Walkhoff’s paste as a filling material in permanentteeth, Ninety-millimeter-diameter petri dishes were filled but long-term treatment success was compromised since with 12 ml of BBL Brain Heart Infusion with PABA the paste eventually resorbed, leaving a deficit in the (BBL Microbiology Systems, Becton Dickinson and Co., root canal. Cockeysville, MD)and BBLPurified Agaro Six-milli- The reported rationales for using Kri 1 paste over meter-diameter wells were created in the agar using ZOEin primary teeth are: ease of use, rapid resorption the end of a sterilized glass pipette. The wells were from periapical tissues, and superior antimicrobial ac- filled with 0.05 ml of either Kri 1 paste or ZOEdis- tion.4,SKri I paste is easier to use than ZOEbecause it is pensed from tuberculin syringes. supplied as a premixed paste that can be placed di- Aliquots of 0.1 ml of 106 CFU/mlStreptococcusfaecalis rectly into the root canal. Kri I resorbs rapidly from the (ATCC19433) in BBLthioglycollate medium were then periapical tissues, however, Woodhouse,et al. 1° have spread over the surface of the agar plates. The plates shown osteolytic changes in the bone surrounding Kri were inverted and incubated at 37°C for 24 hr. Two 1 paste implants in cats, persisting past the retention measurements of the diameter of bacterial growth inhi-

102 Pediatric : March/April 1994 - Volume 16, Number 2 bition around each well were made using a millimeter where T = 51Cr released in test samples, b -- background ruler, and the values were averaged. These values were radiation and R = 5~Cr released in reference samples. then used to calculate area of bacterial inhibition (area For the indirect cytotoxicity evaluation, 0.1 ml of =n d2/4), after subtracting the area of the wells. Agar either Kri 1 paste or ZOEwas placed in the 6.5-ram- plates containing wells, but no medicaments, were diameter chamber insert of a cluster well cell culture spread with S. faecalis and incubated as positive con- plate. The porous bottoms of the cell culture inserts trols. Agar plates containing wells, but no medicaments (uniformly spread 0.4 gm holes) were 1 mmfrom the or S. faecalis, were also incubated to ensure mainte- bottom of the well. This allowed independent access nance of sterile conditions (negative controls). All ex- from the chamber insert into the well through the me- perimental procedures were carried out under sterile dium. One milliliter of 5~Cr-labeled L929cells was dis- conditions using a SterilGARD hood® (The Baker Com- pensed into each cell culture well. The chamber inserts pany, Sanford, ME). containing the test medicaments were then placed into the wells. Empty chamber inserts were placed in the Cytotoxicityevaluation cell culture wells for control. Randomly withdrawn Twoin vitro cytotoxicity test methods were used -- 0.5-ml cell samples were transferred to test tubes to be the direct cell-material contact test of Sp~ngberg17 and used as reference samples for calculating chromium theis indirect cell-material contact test of Safavi, et al. release. The plates containing the cells were incubated Five- to 7-day cultures of L929 mousefibroblast cells at 37°C for 4 hr immediately after mixing (fresh) and were used for the cytotoxicity tests. The culture me- after 1, 7, 14, and 28 days. dium used was minimal essential medium (MEM)with The chamber inserts were stored in PBS and incu- Eagle’s salts (Flow Laboratories, McLean, VA) supple- bated at 37°C and 100 % humidity between experiments. mented with 10% (v/v) fetal calf serum, 2 mM The percentage of 51Cr release in test and control samples glutamine, and 2.2 mg sodium bicarbonate per ml. was calculated by the same formula used in the direct Streptomycin (50 ~g/ml) and penicillin (100 IU/ml) toxicity test. were also added to the medium. The medium was The antimicrobial data were analyzed using changed every other day and on the day before the Student’s t-test while the cytotoxicity data were ana- experiments. lyzed using ANOVA. Radiochromium (SlCr) was supplied as sodium chro- mateTM in sterile isotonic saline (NewEngland Nuclear, Results Du Pont Company, Wilmington, DE). Cell monolayer On average, ZOE-filled wells produced a signifi- culturess were labeled by incubation with I I.tCi per 10 cantly greater area of inhibition (P < 0.01) than did the cells 20-24 hr before the experiments. The labeled cells Kri 1-filled wells (Fig 1). Positive controls showed were harvested using 0.125% trypsin and washed three confluent S. faecalis growth over the agar with no areas times in Ca~÷-free and Mg2÷-free phosphate buffered of inhibition around the wells. Negative controls con- saline (PBS) before use. The cells were resuspended firmed the maintenance of sterile conditions (absence MEMat a concentration of 4x104 cells/ml. of bacterial growth). Serial dilution and plating of the For direct cytotoxicity testing, 0.3 ml of ZOEor Kri 1 paste was placed in the bottom of 16-mm-diameter cell culture wells (Transwell; Costar, ® Cambridge, MA). 400. Materials were tested immediately after mixing (fresh), 350. or after 1-day or 7-day setting times. Twoml of 51Cr- labeled L929 cells were placed in the culture wells and EE 300. incubated in contact with the medicaments for 4 or 24 250. hr at 37°C. After incubation, 1 ml of the fluid in the test chamber 200.: was transferred to a test tube and centrifuged (500 x g) 150, for 10 min. One-half ml of the supernatant was trans- ferred to a test tube for counting in a gammaparticle 100, counter for 1 min. Negative controls of the L929 sus- 50" . pension were used. During cell dispensing, 0.5 ml aliquots of cell suspension were randomly distributed o into six test tubes and used as reference samples. The Kri1 ZOE percentage of SlCr released from the experimental and control samples was calculated according to the for- Medicament mula: Fig 1. Antibacterialaction of Kri 1 paste(N = 15)and ZOE (N 51Cr release (%) = (T-b) x 1 3) basedon Streptococcus faecalis inhibition on agar (mean + (R-b) SD).* = significantlygreater than Kri 1 paste(P < 0.01).

Pediatric Dentistry: March/April 1994 -Volume 16, Number2 103 Both Kri I paste and ZOEwere highly cytotoxic on 4 hr exposure 24 hr exposure lOO--- contact of the fresh medicaments with L929 mouse fibroblast cells (Fig 3) and significantly different from the control (P < 0.01). There was no significant differ- ence between ZOEand control in the 1-, 7-, 14-, and 28- day experiment groups. The cytotoxicity of Kri I paste was statistically higher than ZOEand the control in the ~4o2 302 1- and 7-day experimental groups (P < 0.01). Between and 14 days, the cytotoxicity of Kri I paste dropped to

lO: a level that was not significantly different from either 0: ZOEor the control and remained at this level to termi- SettingTime (days) nation of the experiment at 28 days. Discussion Fig 2. Relativetoxicity of Kri I pasteand ZOE based on percent ~lCr releasedin the direct contactexperiments (mean + SDof The finding that ZOEproduced better antimicrobial six experiments).Control values have been subtracted from activity than Kri 1 paste contrasts with several other experimentalvalues. * = significantlygreater than ZOE (P studies showing Kri 1 paste to be the superior antimi- o.ol). crobiaU4, ~s The increased antimicrobial action of ZOE in this study, however, mayreflect the increased amount bacterial suspension confirmed a bacterial concentra- of eugenol incorporated into the zinc oxide to approxi- tion of 1x106/ml. mate the consistency of Kri 1 paste. The limitations of Four-hour exposure of the mouse fibroblast cells to Kri 1 paste resulted in a high release of 51Cr, whether the method of antimicrobial testing employed in this study and in the studies of Orstavik 14 and Seow~s must fresh, at I day, or at 7 days (Fig 2). This SlCr release was be recognized. The results may represent a variation in significantly greater than the controls (P < 0.01). Four- the diffusion rate of the medicaments through the agar hour exposure of the cells to fresh, 1-, and 7-day ZOE or the effectiveness of a particular medicament in a resulted in 51Cr release that was significantly greater gaseous phase2 A better assessment of antimicrobial than the controls (P< 0.01). Although fresh ZOEre- action such as the methodsof Sp&ngberg,et al. 19 or Hill, sulted in a high release of S~Cr (approximately equal to et al., 2° in which antimicrobial action is determined by that of Kri I paste), the release rates after I and 7 days the action of serial dilutions of an antibacterial solution decreased to levels that were significantly lower than on a test organism, was not possible in this study since Kri I paste (P < 0.01). neither Kri I paste or ZOEformed a homogenoussolu- Twenty-four-hour exposure of the cells to Kri I paste tion in an appropriate solvent. and ZOEalso resulted in a high release of S~Cr, inde- S. faecalis was used in this experiment because it has pendent of the setting time. The release of ~lCr for both Kri I paste and ZOEwas significantly greater than the been shown to occur in pulpal infections in permanent teeth 21, ~ and, as a group, the betahemolytic strepto- controls (P < 0.01). The cytotoxicity of Kri I paste was cocci have been shown to infect the primary root ca- significantly greater than ZOEat all setting times in the nal. a~-2s In addition, the organism is easy to grow and it 24-hr exposure experiments (P < 0.01). has been used extensively in experimental antiseptic susceptibility studies in . Most importantly,

80. S. faecalis efficiently and rapidly colonizes dentinal tu- bules compared with other species 26 and therefore is 70 particularly difficult to eradicate. It represents a stan- 6O dard against which the antimicrobial action of a medi- 50 cament should be tested. 40 The 5~Cr release method of cytotoxicity testing is a standard biocompatibility test, accurate and sensitive in assessing cell injury. 27 L929mouse fibroblast cells are ~o commonlyused in biocompatibility tests and were used in this study for comparisonof cytotoxicity results from SettingTime (days) other studies. The 7-day set used in this study is an acceptable Fig 3. Relativetoxicity of Kri I pasteand ZOE based on percent maximumin the direct exposure experiments because one week is considered sufficient for any material to ~lCrreleased in the indirectcontact experiments (mean + SDof 27 six experiments).Control values have been subtracted from reach optimal setting. These time points were chosen experimentalvalues. * = significantlygreater than ZOE (P to investigate the initial cytotoxicity of the medica- 0.01). ments and the time period over which they might lose

104 Pediatric Dentistry: March/Aprll 1994 - Volume 16, Number 2 their cytotoxic effect. The cells were exposed to the cells in direct contact with either Kri 1 paste or ZOE ZOEand Kri I paste for 4 and 24 hr as prescribed in the may be killed, while PDL cells not directly in contact ANSI/ADA document No. 41o 28 Four hours has been may be affected to a greater degree by Kri I paste than shown to be sufficient time for the cells to attach (under ZOE. normal conditions) and to react to a material. 29 The 24- hr contact time was chosen to detect the effects of pro- Conclusion longed exposure to the material. The SlCr release value This study indicates Kri I paste is less effective than recorded for the L929 mouse fibroblast cells after a 4-hr ZOEin its antibacterial activity. Kri I paste is at least as contact with fresh medicaments in the direct contact cytotoxic as ZOE on cells in direct contact with the experiments was expectedly high for both Kri 1 paste medicament and is more cytotoxic than ZOE on cells and ZOE. The eugenol in the ZOE and the iodoform, not in direct contact. camphor, p-chlorophenol, and menthol in Kri I paste, Dr. Wrightis fellow, departmentof pediatric dentistry and orthodon- are free to interact and exert their cytotoxic effect upon tics; Dr. Barbosais visiting professor, departmentof restorative den- the mouse fibroblast cells. The significant decrease in tistry and endodontology;Dr. Arakiis visiting professor, department of restorative dentistry and endodontology;and Dr. Sp~ngbergis ZOE cytotoxicity after 1 and 7 days, over which time head, departmentof restorative dentistry and endodontology;all at the toxicity of Kri 1 paste remained high, may be ex- University of ConnecticutHealth Center, Farmington,Connecticut. plained by the fact that ZOEsets while Kri I paste does The authors thank Mr. David O’Sullivan and Drs. DuaneDouglas and not. The eugenol in set ZOE is bound and mostly un- NormanTinanoff for editorial assistance. available to exert its cytotoxic effect. 1. CoxST, HembreeJH Jr, McKnightJP: The bactericidal potential Although there were significant differences between of various endodontic materials for primary teeth. Oral Surg Kri I paste and ZOEtoxicities in the 24-hr direct contact Oral MedOral Pathol 45:947-54, 1978. 2. experiments, the high levels of slCr release for both Kri Goerig AC, CampJH: in primary teeth: a review. Pediatr Dent 5:33-7, 1983. 1 paste and ZOE indicate the medicaments are equally 3. CampJH: Pulp therapy for primary and young permanent cytotoxic when fresh, and after 1 and 7 days. These teeth. Dent Clin North Am28:651~8, 1984. results differ from the 4-hr direct contact experiments 4. Garcia-GodoyF: Evaluation of an iodoformpaste in root canal and indicate no difference in cytotoxicity of the two therapy for infected primaryteeth. ASDCJ Dent Child 54:30-4, 1987. medicaments if they are allowed to contact the target 5. Rifkin A: A simple, effective, safe techniquefor the root canal cells for a long enough period of time. The difference treatment of abscessed primary teeth. ASDCJ Dent Child between the cytotoxicity of Kri 1 paste and ZOE -- 47:435-41,1980. expressed as a higher release of SlCr caused by Kri 1 6. Rifkin A: The root canal treatment of abscessedprimary teeth-- paste -- may be due to the greater toxicity of the indi- a three to four year follow-up. ASDCJ Dent Child 49:428-31, 1982. vidualB° constituents of Kri I paste. 7. Castagnola L, Orlay HG:Treatment of gangrene of the pulp by The indirect method of cytotoxicity testing provides the Walkhoffmethod. Br Dent J 93:93-102, 1952. an additional assessment of material cytotoxicity and 8. Juge H: Resorbablepastes for root canal fillings. Int Dent J loss of cytotoxic effect with time. The decrease in ZOE 9:461-76, 1959. cytotoxicity to control levels at 1-day set and beyond in 9. Barker BCW,Lockett BC: Endodontic experiments with resorbable paste. AustDent J 16:364-72,1971. the indirect contact experiments may be explained by 10. Woodhouse BM, Savage NW, Monsour FNT: Radiographic the setting of this material in the same manner as ex- evaluation of intraosseous implants of endodontic materials. plained above for the direct contact experiments. The Oral Surg Oral MedOral Pathol 71:218-22, 1991. dramatic decrease in the cytotoxicity of ZOE in this 11. ErausquinJ, Muruz~balM: Root canal fillings with zinc oxide- study, however, implies that over the first few days, eugenol cement in the rat molar. Oral Surg Oral MedOral Patho124:547-58,1967. direct contact is a more important aspect of ZOE cyto- 12. BrowneRM, Friend LA:An investigation into the irritant prop- toxicity than is the set of the material. As expected, the erties of someroot filling materials. ArchOral Bio113:1355-70, cytotoxicity of Kri 1 paste remains high for a long pe- 1968. riod of time. The decline to control levels by 14 days 13. Friend LA, BrowneRM: Tissue reactions to someroot filling despite the lack of a set state, may occur due to leaching materials. Br Dent J 125:291-98,1968. 14. Orstavik D: Antibacterial properties of root canal sealers, ce- of toxic constituents of the medicament into the PBS. ments and pastes. Int EndodJ 14:125-33, 1981. Browne and Friend’s implant studies in rabbits 12’ 13 15. SeowWK: The effects of dyadic combinations of endodontic suggest the in vivo effects of Kri 1 paste and ZOE are medicaments on microbial growth inhibition. Pediatr Dent represented better by the indirect experiment rather 12:292-97,1990. 16. MeryonSD, BrookAM: In vitro comparisonof the cytotoxicity than the direct experiments. Kri 1 paste reportedly of twelve endodontic materials using a new technique. Int caused considerable tissue necrosis as an immediate EndodJ 23:203-10,1990. response compared with a small zone of tissue necrosis 17. Sp~ngbergL: Kinetic and quantitative evaluation of material localized to the material surface caused by freshly mixed cytotoxicity in vitro. Oral Surg Oral MedOral Pathol 35:389- ZOE, with the amount of ZOE-induced necrosis de- 401, 1973. 18. Safavi KE, SpSngbergLSW, Costa NS Jr, Sapounas G: An in pendent on the set of the ZOEo12,1Bina clinical situation, vitro method for longitudinal evaluation of toxicity of it might be expected that periodontal ligament (PDL) endodontic sealers. J Endod15:484-86, 1989.

Pediatric Dentistry: March/April1994 -Volume 16, Number2 105 19. Sp~ngbergL, Sa favi KE, KaufmanA, Pascon EA: Antimicrobial 25. Tomic-KarovicK, Jelinek E: Comparativestudy of the bacterial and toxic effect in vitro of a bisdequaliniumacetate solution for flora in the surroundings, the root canals, and sockets of de- endodontic use. J Endod14:175-78, 1988. ciduous molars. Int Dent J 21:375-78,1971. 20. Hill SD, Berry CW,Seale NS, Kaga M: Comparisonof antimi- 26. Orstavik D, HaapasaloM: Disinfection by endodontic irrigants crobial and cytotoxic effects of glutaraldehydeand formocresol. and dressings of experimentally infected dentinal tubules. Oral Surg Oral MedOral Pathol 71:89-95, 1991. EndodDent Traumatol 6:142-49, 1990. 21. Engstr6mB: The significance of enterococci in root canal treat- 27. Pissiotis E, Sp~ngbergLSW: Toxicity of Pulpispad using four ment. Odontol Revy 15:87-106, 1964. different cell types. Int EndodJ 24:249-57,1991. 22. GoldmanM, Pearson AH:Postdebridement bacterial flora and 28. AmericanNational Standards Institute: DocumentNo. 41. Rec- antibiotic sensitivity. Oral Surg Oral MedOral Pathol 28:897- ommendedStandard Practices for Biological Evaluation of 905, 1969. Dental Materials. AmericanNational Standards Institute, New 23. CohenMM, Joress SM,Calisti LP: Bacteriologic study of in- York, 1982. fected deciduous molars. Oral Surg Oral Med Oral Pathol 29. SpangbergL, LangelandK: Biologic effects of dental materials. 13:1382-86,1960. 1. Toxicityof root canal filling materials on HeLacells in vitro. 24. MarshSJ, Largent MD:A bacteriologic study of the pulp canals Oral Surg Oral MedOral Pathol 35:402-14, 1973. of infected primary molars. ASDCJ Dent Child 34:460-70, 30. Gleason MN,Gosselin RE, HodgeHC, Smith RP: Clinical Toxi- 1967. cology of CommercialProducts. 3rd ED. The Williams and Wilkins Co. Baltimore, 1969.

AAPDsalutes the consultants for the 1994 American Board Section Examinations George Acs - Pleasantville, New York Steven M. Adair - Augusta, Georgia Robert A. Boraz - Kansas City, Kansas Steven K. Brandt - Temple, Texas Charles M. Brenner - Rochester, New York Paul S. Casamassimo - Columbus, Ohio James A. Croll - York, Pennsylvania Joseph C. Creech, Jr. - Mesa, Arizona Diane C.H. Dilley - Chapel Hill, North Carolina Kevin J. Donly - Iowa City, Iowa Robert R. Gatehouse - Middletown, Connecticut Stephen J. Goepferd - Iowa City, Iowa Norman L. Goldberg - Concord, Massachuesetts David L. Good - Conoga Park, California Paul D. Herer - Vernon Hills, Illinois Margot H. Jaffe - New York, New York Paul E. Kittle - Fort Leavenworth, Kansas Doron Kochman - Pittsford, New York Donald W. Kohn - New Haven, Connecticut Larry S. Luke - Los Angeles, California William A, Mueller - Denver, Colorado John E. Nathan - Oak Brook, Illinois M.M. Nazif - Pittsburgh, Pennsylvania Peter W. H. Ngan - Columbus, Ohio Eduardo A. Ovadia - Mexico City, Mexico John E. Peterson, Jr. - LomaLinda, California Rolando A. Prado - Wilson, North Carolina James W. Preisch- - Columbus, Ohio Indru C. Punwani - Evanston, Illinois Michael W. Roberts - Chapel Hill, North Carolina Peter J. Ross - Lancaster, Pennsylvania HowardS. Schneider - Jacksonville, Florida Robert S. Sears - Herndon, Virginia Preston G. Shelton - Baltimore, Maryland Andrew L. Sonis - Newton Highlands, Massachusetts Robert H. Spedding - Lexington, Kentucky Robert L. Stonerock - Federal Services George P. Thomas o Silver Spring, Maryland Erwin G. Turner - Lexington, Kentucky Richard D. Udin - Hawthorne, California Paul L. Vitsky - Fredericksburg, Virginia William F. Waggoner - Columbus, Ohio Paul O. Walker - Dallas, Texas James A. Weddell, Indianapolis, Indiana Paul B. Wiegand - San Antonio, Texas Ronald L. Winder - Tulsa, Oklahoma

106 PediatricDentistry: March/April 1994 - Volume16, Number2