S. Haenni, P. R. Schmidlin, B. Mueller, B. Sener & M. Zehnder
Division of Endodontology, Department of Preventive Dentistry, Cariology, and Periodontology, Center for Dental Medicine, University of Zurich, Zurich, Switzerland.
Swiss Federal Institute for Environmental Science and Technology (EAWAG), Kastanienbaum, Switzerland.
Swiss Federal Institute of Technology (ETH), Limnological Research Center, Kastanienbaum, Switzerland.

Introduction.
Teeth with periapical lesions of endodontic origin may be divided into two main groups: teeth with necrotic pulps (primary endodontic infections) and teeth with failed root canal treatments. In cases of primary endodontic infections, the flora consists mostly of anaerobes (Sundqvist 1976). In such cases, a calcium hydroxide (Ca(OH)2) dressing for at least 1 week has been demonstrated to kill bacteria more thoroughly than all other common intracanal medications (Bystrom et al. 1985), and it has been recommended for the treatment of apical periodontitis in teeth with necrotic pulps. However, in cases of failed endodontic treatment, the intracanal flora is different, and facultative anaerobes may predominate (Molander et al.1998). In addition, yeasts have been associated with endodontic failures (Waltimo et al. 1997). Interestingly, Ca(OH)2 has been shown to be inefficient in the killing of both facultative anaerobes and yeasts, whilst other medications or irrigating solutions have been shown to be more effective in the killing of these microbiota in vitro (Crstavik & Haapasalo1990,Waltimo et al.1999). It has therefore been suggested that Ca(OH)2 powder should be mixed with iodine potassium iodide (IPI), chlorhexidine (CHx) or sodium hypochlorite (NaOCl) in order to obtain a wide-spectrum antimicrobial preparation with a long-lasting effect (Waltimo et al. 1999). However, the effects of these irrigation solutions on Ca(OH)2 and vice-versa have not been studied in detail.
When used as an intracanal dressing, Ca(OH)2 mixed with water or saline to a paste-like consistency results in a radially diminishing pH rise in dentine over time, until a steady state is reached (Tronstad et al.1980). Alkalis have a pronounced destructive effect on cell membranes and protein structure (Gordon et al. 1985), and antimicrobial properties of Ca(OH)2 are directly related to pH (Bystrom et al. 1985, Evans et al. 2002). Therefore, the purpose of this study was to evaluate the effect of mixing Ca(OH)2 powder with IPI, CHx digluconate and NaOCl solutions on the ability of Ca(OH)2 to raise the pH at the root surface in vitro. In addition, the antimicrobial potential of these solutions and mixtures was assayed with an agar diffusion test against Enterococus faecalis and Candida albicans.

Materials and methods.

Selection and preparation of teeth.
Eighty single-rooted human canines and premolars of similar length and diameter from the Department’s collection of extracted teeth were used. After collection, they were stored in 0.1% thymol solution. Crowns were amputated at the level of the cemento-enamel junction. Root canals were instrumented with ProTaper1 files (Dentsply Maillefer, Ballaigues, Switzerland) under copious irrigation with a1% NaOCl solution. The apices were intentionally over instrumented witha size 40Flexofi le (Maillefer) to facilitate sealing with glass ionomer cement (see below). After instrumentation, the root canals were flushed repeatedly with17% EDTA solution for 5 min to remove the smear layer. In the middle third of the buccal aspect of each root, a standardized well 0.75 mm deep, 3 mm long and 1.5 mm in diameter was prepared with a 1.5-mm diameter diamond round bur. To assess the remaining dentine thickness under the measuring well, digital radiographs were taken of the mesio-distal aspect of the teeth (Digora1, Soredex, Helsinki, Finland). After calibration, the dentine thickness was measured using the Digora1 software (Soredex). Before the experiments, teeth were washed in deionized water at room temperature for12 h.

Intracanal dressings.
The 80 experimental teeth were randomly divided into four different test groups and four corresponding control groups of 10 teeth each. The teeth were placed into individual vials containing unbuffered isotonic saline solution. After air-drying the teeth and sealing the apices with glass ionomer cement (Ketac ESPE, Seefeld, Germany) and nail varnish, intracanal medications were applied. Prepared root canals were filled with: Group 1, Ca(OH)2 (Merck, Darmstadt, Germany) mixed with unbuffered isotonic saline; Group 2, unbuffered isotonic saline; Group 3, 0.5% (w/v) CHx digluconate Ca(OH)2 paste; Group 4, 0.5% CHx; Group 5, 1% (w/v) NaOCl Ca(OH)2 paste; Group 6, 1% NaOCl; Group 7, 5% (w/v) IPI (I2 ю KI) Ca(OH)2 paste and Group 8,5%IPI.According to clinical standards, all pastes were mixed to a creamy consistency on a glass slab using Ca(OH)2 powder and the respective solutions (1 :1.5, w/v). Pastes were applied with a lentulo spiral (Maillefer), solutions with a syringe and a 26-gauge needle. Subsequently, access cavities were sealed in the same manner as the apices. Teeth were processed in batches containing one sample of each test (mixtures) and control (solutions only) group. To further avoid bias, the investigator taking the pH measurements was not aware of the content of the roots.

pH measurements.
Measurements of root surface pH were performed immediately after filling and then after 24 h, 3 days, 1, 2, 3 and 5 weeks postfill. For measuring, teeth were removed from their vials, cooled to room temperature in a 100% humid environment and dried with compressed air (GEPE-air, Image Trade, Safenwil, Switzerland). Two microlitres of unbuffered isotonic saline solution was placed into the standardized well. After an equilibration time of 5 min, pH was measured with a calibrated microelectrode (MI 415-2, Microelectrodes Inc., Bedford, NH, USA). Subsequently, teeth were returned to their vials filled with fresh saline and incubated at 378C.
In addition, the pH of 1:1.5 (w/v) mixtures of Ca(OH)2 with the irrigating solutions was measured with a semiconductor pH electrode (Sanwa Tsusho Co., Tokyo, Japan) and compared to the pH of the solutions (Table 1).

Table 1. pH values of irrigants/Ca(OH)2 medications and pH values measured at the root surface of teeth dressed with these in vitro at different points in time.

To assess the influence of removing the cementum layer on the experimental tooth surface (measuring well) on hydroxide ion (OH-) penetration, teeth were stained with a pH indicator (thymol blue, Merck). Thymol blue changes its colour from yellow to blue at a pH range of 8.0-9.2. Experimental and control teeth that had been dressed for 5 weeks were unsealed, and the root canals cleaned with saline and interproximal brushes. Teeth were then placed in eppendorf tubes containing 1mL of the indicator solution. Subsequently, specimens were centrifuged at 4000 g for 30 min, and longitudinally or horizontally sectioned through the centres of the measuring wells using a slow-speed diamond-coated rotary disc (Isometh, Buehler, Prufmaschinen AG, Zurich, Switzerland).The disc was cooled with kerosene to avoid influencing pH scores, and finally the cut surfaces were photographed.

Agar diffusion test.
Facultative bacteria (E. faecalis ATCC 29212) and yeasts (C. albicansOMZ110) were maintained separately influid universal medium (FUM) (Gmur & Guggenheim 1983). The latter microbiota were chosen because they are the facultatives and yeasts most frequently isolated from root canals of failed treatments (Waltimo et al.1997, Dahle. n et al. 2000). The optical density of FUM sample aliquots was measured at a wavelength of 550 nm in a spectrophotometer (U 2000, Hitachi, Boehringer-Mannheim, Rotkreuz, Switzerland) and adjusted by dilution with FUMto1.0. In Petri plates,25 mL of Columbia Blood Agar (Becton Dickinson Microbiology Systems, Sparks, MD,USA)was inoculatedwith5 mL of the broth containing the microorganisms. After the agar had set, round cavities, 5 mm in diameter, were punched out. These cavities were filled with the solutions (40 mL) and combinations (40 mL of solution mixed with 20 mg Ca(OH)2 powder). After 24 h incubation at 378C in ambient air, diameters of zones of inhibition were measured with a caliper. All experiments were performed in triplicate.
To test the influence of high OH- ion concentration on the antimicrobial effectiveness of irrigants, 5% IPI and 0.5% CHx solutions were adjusted to pH 12 by titration with a 1% sodium hydroxide (NaOH) solution (pH 14), and subjected to the agar diffusion test described above.

Statistical analysis.
As pH values are logarithmic, nonparametric statistics were applied to compare pH values at the root surface between groups: Kruskal-Wallis one-way analysis of variance followed by Mann-Whitney U-test for individual comparisons. Data obtained in the agar diffusion tests and dentine thickness values were compared using one-way analysis of variance (anova) followed by an unpaired t-test. Bonferroni adjustments were applied for multiple posthoc testing. Levels for rejection of null hypotheses were set at P < 0.05.

Results.

pH measurements.
High pH values in the range of11.7-12.4 were measured in suspensions of Ca(OH)2 powder with irrigating solutions, irrespective of the pH of the irrigating solution alone (Table 1). In the extracted single-rooted teeth used for the OH-ion diffusion assay, the average dentine thickness between the root canal space and the floor of the measuring well was1.8 _0.3 mm. No significant difference in dentine thickness was found between groups (anova, P = 0.76). Removing the cementum layer at the root surface to obtain measuring well did not influence OH- ion penetration (Fig. 1). Significant pH differences at the root surface between test and control groups were first noted after 2 weeks (Table 1; Kruskal-Wallis test, P < 0.05). These differences became more pronounced at weeks 3 and 5 (P < 0.005). Significant differences were found from week 2-5 (Mann-Whitney U-test, P < 0.05) when comparing pH values of the IPI/Ca(OH)2 group and the saline/Ca(OH)2 group with pH values obtained with their corresponding irrigation solutions. It took 3 weeks for the CHx/Ca(OH)2 group and 5 weeks for the NaOCl/Ca(OH)2 group to produce significant pH differences compared to controls (P < 0.05). There was a tendency of the NaOCl/Ca(OH)2 medication to release less OH- ions through dentine than the other Ca(OH)2/ irrigant mixtures. However, no significant differences between pH values measured with different Ca(OH)2 medications were noted at any time (P > 0.05). This statement was also true for pH differences between the four control groups.

Figure 1. Teeth dressed for 5 weeks with a calciumhydroxide (Ca(OH)2) medication (panels B and C) or saline (panel A) were perfused with thymol blue (original magnification 30x). The indicator changes its colour from yellow to blue at pH 8.0^9.2. Note that the removal of cementum in the measuring well apparently did not influence the penetration of hydroxide (OH-) ions through dentine (panels B and C).

Agar diffusion test.
On plates incubated with E. faecalis or C. albicans, zones of inhibition did not differ significantly between the four Ca(OH)2 medications (Table 2; anova, P > 0.05). Significantly larger areas of inhibition were obtained with CHx alone than with the same amount of CHx mixed with Ca(OH)2. Inhibit on areas with I2/I_ as well as with OCl- were significantly smaller than respective zones obtained with the corresponding mixtures with Ca(OH)2 (P < 0.05).
When the pH of the 0.5% CHx solution was increased to 12 by titration with NaOH, the agent precipitated. However, the solution was still as effective against C. albicans as the original at pH 6 (P > 0.05). In contrast, a 5% IPI solution made alkaline with NaOH turned clear at pH 10.9 and did no longer inhibit the growth of C. albicans.

Table 2. Diameters (mm) of the zones of inhibition against the test organisms (agar diffusion).

Discussion.
The antimicrobial effectiveness of Ca(OH)2 is based on its ability to release OH- ions (Proell1949). Because Ca(OH)2 is a strong base, weak acids in a Ca(OH)2 suspension do not influence chemical or antimicrobial properties of the latter. This was corroborated by the present study.
In this in vitro study, similar pH values were measured at the root surface as in an in vivo investigation in monkeys (Tronstad et al. 1980). In the absence of fluid movement in the nonvital tooth, OH- ions are transported through dentine tubuli by molecular diffusion (Nerwich et al.1993). Thus, the development of the pH in a defined solution volume outside the tooth can be calculated using Fick’s first law of diffusion (Crank1975). Starting with a pH of 12 at the Ca(OH)2-dentine interface, a diffusion length of 1.8 mm and an average liquid space of 0.5% in root dentine (Pashley 1991), a calculated pH of 9.6 would be expected after 5 min in a volume of 2 mL when all chemical interactions are neglected. However, the median pH value at the surface of roots filled with Ca(OH)2 pastes measured after 3 weeks was 8.7. Thus, the reported buffering capacity of dentine for alkalis (Wang&Hume1988) was confirmed in this study.This buffering capacity may also explain why NaOCl alone, at a similar pH as the Ca(OH)2 pastes, did not change the root pH in our experiments. Ca(OH)2 has low solubility in water. In the moist environment of the root canal system, undissolved Ca(OH)2 in a paste-like suspension will steadily dissolve, resulting in a sustained pH effect, which was not observed with the solutions alone. This is in agreement with a reported in vivo finding: a saturated solution of Ca(OH)2, pH 12,wasunable to eliminate E. faecalis fromthe root canal of a cat canine, whilst a paste-like aqueous suspension killed all cultivable bacteria (Stevens & Grossman1983).
As stated above, Ca(OH)2 effectiveness is entirely pH related (Bystrom et al.1985, Evans et al. 2002).Although direct clinical evidence is still elusive, it may therefore be stated that bacteria and yeasts with a high tolerance for basic pH levels, such as Enterococci and Candida species, may not effectively be killed by conventional Ca(OH)2 suspensions. In particular, Enterococci are hardy opportunistic invaders, able to survive as a mono infection in a sparse environment (Fabricius et al. 1982). Moreover, Enterococci readily invade dentinal tubules by division (Crstavik & Haapasalo1990). Therefore, facultative anaerobes and yeasts may find an ecological niche in dentinal tubules of root-treated teeth. In endodontic retreatments, when facultatives predominate, there appears to be a need for improved interappointment root canal dressings. Under the conditions of this study, no increase of antimicrobial efficacy was noted in suspensions of irrigating solutions with Ca(OH)2 compared to a conventional Ca(OH)2 medication. Our results are in accordance with a recently published study (Estrela et al. 2001), which could not demonstrate an additive antibacterial effect when mixing Ca(OH)2 powder with CHx, but showed that Ca(OH)2 does not lose its antibacterial properties in such a mixture. In the original article on mixtures of root canal disinfectants (Waltimo et al. 1999), irrigating solutions were saturated with Ca(OH)2 powder. In those mixtures, a saturated Ca(OH)2 solution alone was less effective against C. albicans than NaOCl, IPI or CHx solutions saturated with Ca(OH)2 powder in vitro. However, as stated above, a saturated Ca(OH)2 solution is ineffective as a root canal dressing in vivo (Stevens & Grossman 1983). Therefore, it may be more relevant to use Ca(OH)2/irrigant mixtures with a paste-like consistency for comparative in vitro studies.
An agar diffusion method was chosen in the current study to assess the antimicrobial effectiveness of irrigants and their mixtures with Ca(OH)2. It may be argued that, with the known inhibitory action of dentine on root canal medicaments (Haapasalo et al. 2000), a dentine block model may have been more appropriate. However, the primary goal of the present investigation was to study the direct interactions of irrigants and Ca(OH)2, and an agar diffusion test appeared suitable for this purpose.
In CHx/Ca(OH)2 mixtures, we detected an inhibition of antimicrobial activity of CHx, most likely caused by the high pH and the alkaline buffering capacity of the suspension. The reason for the reduced efficacy may be the deprotonation of the biguanide at pH >10 and hence a markedly reduced solubility and altered interaction with bacterial surfaces due to the change in the charge of the molecule (Jones et al.2000). Ina 0.5%CHx digluconate solution inwhichthepHwasadjustedto12byNaOH titration, the antimicrobial effect was maintained. An explanation for that phenomenon may be seen in the lower alkaline buffering capacity of a solution compared to a Ca(OH)2 suspension, resulting in a pH drop with increasing distance from the OH- source. Dilution and the buffering effect of the agar may have allowed the CHx to become reprotonated and regain its bactericidal activity. IPI appears to be ineffective in killing microbiota at pH levels higher than11. I2 at high pH disproportionates to IO_ and I_ and, since recombination is kinetically slow, looses its antimicrobial activity (Lengyel et al. 1993). In addition, the allergenic potential of iodine may be a contraindication for its long-term use as a root canal dressing.
Thus, the most promising Ca(OH)2 powder/irrigant combination might be the Ca(OH)2/NaOCl paste. Hypochlorite is chemically stable at high pH. Although apparently not more effective in an agar diffusion test, such a combination may have better tissue-dissolving properties than conventional Ca(OH)2/saline paste (Hasselgren et al.1988). Further research is necessary to confirm that hypothesis.

References.

Bystrom A, Claesson R, Sundqvist G (1985) The antibacterial effect of camphorated paramonochlorophenol, camphorated phenol and calcium hydroxide in the treatment of infected root canals. Endodontics and DentalTraumatology1,170-5.
Crank J (1975) The Mathematics of Diffusion, 2nd edn. Oxford: Clarendon Press.
Dahle. n G, Samuelsson W, Molander A, Reit C (2000) Identification and antimicrobial susceptibility of Enterococci isolated from the root canal. Oral Microbiology and Immunology 15, 309-12.
Estrela C, Bammann LL, Pimenta FC, Pecora JD (2001) Control of microorganisms invitro by calciumhydroxide pastes. International Endododontic Journal 34,341-5.
Evans M, Davies JK, Sundqvist G, Figdor D (2002) Mechanisms involved in the resistance of Enterococcus faecalis to calcium hydroxide. International Endododontic Journal 35, 221-8.
Fabricius L, Dahle. n G, Holm SE, Moller AJ (1982) Influence of combinations of oral bacteria on periapical tissues of monkeys. Scandinavian Journal of Dental Research 90, 200-6.
Gmur R, Guggenheim B (1983) Antigenic heterogeneity of Bacteroides intermedius as recognized by monoclonal antibodies. Infection and Immunity 42, 459-70.
Gordon TM, Ranly DM, Boyan BD (1985) The effects of calcium hydroxide onbovine pulptissue: variations in pH and calcium concentration. Journal of Endodontics 33,126-31.
Haapasalo HK, Sire. n EK, Waltimo TMT, Crstavik D, Haapasalo MPP (2000) Inactivation of local root canal medicaments by dentine: an in vitro study. International Endododontic Journal 32, 421-9.
Hasselgren G, Olsson B, Cvek M (1988) Effects of calciumhydroxide and sodium hypochlorite on the dissolution of necrotic porcine muscle tissue. Journal of Endodontics14,125-7.
Jones DS, Brown AF, Woolfson AD, Dennis AC, Matchett LJ, Bell SE (2000) Examination of the physical state of chlorhexidine within viscoelastic, bioadhesive semisolids using raman spectroscopy. Journal of Pharmacological Science 89,563-71.
Lengyel I, Epstein IR, Kustin K (1993) Kinetics of iodine hydrolysis. Inorganic Chemistry 32,5880-2.
Molander A, Reit C, Dahle. nG, Kvist T (1998) Microbiological status of root-filled teeth with apical periodontitis. International Endododontic Journal 31,1-7.
Nerwich A, Figdor D, Messer HH (1993) pH changes in root dentin over a 4-week period following root canal dressing with calcium hydroxide. Journal of Endodontics 19,302-6.
Crstavik D, Haapasalo M (1990) Disinfection by endodontic irrigants and dressings of experimentally infected dentinal tubules. Endodontics and Dental Traumatology 6,142-9.
Pashley DH (1991) Clinical correlations of dentin structure and function. Journal of Prosthetic Dentistry 66,777-81.
Proell F (1949) Ueber die Eigenschaften des Calxyls und seine Vorzugevorandereninder zahnarztlichenPraxisangewandten Medikamenten. Zahnarztliche Rundschau14, 255-9.
Stevens RH, Grossman LI (1983) Evaluation of the antimicrobial potential of calcium hydroxide as an intracanal medicament. Journal of Endodontics 9,372-4.
Sundqvist G (1976) Bacteriological Studies of Necrotic Dental Pulps. Sweden: UmeP, UmeP University.
Tronstad L, Andreasen JO, Hasselgren G, Kristerson L, Riis I (1980) pH changes in dental tissues after root canal filling with calcium hydroxide. Journal of Endodontics 7,17-21.
Waltimo TM, Crstavik D, Sire. n EK,Haapasalo MP (1999) In vitro susceptibility of Candida albicans to four disinfectants and their combinations. International Endododontic Journal 32, 421-9.
Waltimo TM, Sire. n EK,Torkko HL, Olsen I, Haapasalo MP (1997) Fungi in therapy-resistant apical periodontitis. International Endododontic Journal 30,96-101.
WangJD, HumeWR (1988) Diffusion of hydrogen ion and hydroxyl ion from various sources through dentine. International Endododontic Journal 21,17-26.