Ultrastructural Examination of Failed Molar …Ultrastructural Examination of Failed Molar...

Case Report/Clinical Techniques

Ultrastructural Examination of Failed Molar Retreatmentwith Secondary Apical Periodontitis: An Examination ofEndodontic Biofilms in an Endodontic Retreatment FailureGary B. Carr, DDS,* Richard S. Schwartz, DDS,

†Christoph Schaudinn, PhD,

‡Amita Gorur, MSc,

§

and J. William Costerton, PhDk¶

Abstract
A light and electron microscope examination of the re-sected root tip of a failing endodontically re-treatedlower molar was examined. The tooth had been initiallytreated 10 years ago and then re-treated 2 years ago.The resected root tip was sectioned axially, and thinsections were examined through the entire length ofthe specimen. Thin sections were examined with a trans-mission electron microscope. The thin sections wererandomly chosen along the isthmus areas between themesiobuccal and mesiolingual canals. Our findingssuggest that a complex, variable, multispecies biofilmwas present the entire length of the specimen. (J Endod2009;35:1303–1309)
Key WordsApical periodontitis, biofilm, endodontic failure,endodontic retreatment, ultrastructural

From *Pacific Endodontic Research Foundation, San Diego,CA; †Private practice, San Antonio, TX; ‡Center for Biofilms,USC School of Dentistry, Los Angeles, CA; §Lawrence BerkeleyNational Laboratory, Berkeley, CA; kDepartment of OrthopedicSurgery, Allegheny General Hospital, Pittsburgh, PA; and¶Center for Genomic Sciences, Allegheny-Singer ResearchInstitute, Pittsburgh, PA.

Address requests for reprints to Dr Gary B. Carr, PacificEndodontic Research Foundation, 6235 Lusk Blvd, San Diego,CA 92121. E-mail address: [email protected]/$0 - see front matter

Copyright ª 2009 American Association of Endodontists.doi:10.1016/j.joen.2009.05.035

JOE — Volume 35, Number 9, September 2009

Primary apical periodontitis is one of the most common bacterial diseases ofhumans, afflicting as many as 60% of all individuals worldwide at some point

in their life span (1). It has been established conclusively that the cause of thedisease stems from the microbial colonization of the root canal space by oralmicroorganisms, leading to subsequent periapical breakdown from the spread ofthese organisms, or their toxins, into the periodontal ligament space and bone(2–4). Of the more than 700 species of bacteria thought to be present in the mouth(5), it appears only a limited number of species are capable of surviving and per-sisting in the root canal environment, and an even smaller number are capable ofsurviving endodontic treatment regimens whose sole purpose is directed towardeliminating them (6–10).

Contemporary research has also tried to make a distinction between primaryapical periodontitis and secondary apical periodontitis (6–12). The distinctionproposed is that in primary apical periodontitis (disease present before any treatment)the organisms are predominately gram-negative obligate or facultative anaerobes thatare multispecies in character, whereas in secondary disease (post-treatment disease)the organisms are facultative gram-positive organisms of a very limited species distri-bution and are perhaps even monoinfections (11–13).

There is much recent work that demonstrates that the most likely disease modelfor endodontics is a biofilm model, just as it is in periodontal disease, and that althoughacute disease might be caused by planktonic organisms, the source for all these organ-isms is from a preexisting and ever-present biofilm (14–19). By using combined lightand electron microscopy, Nair et al (18–20) have shown conclusively that in bothprimary and secondary apical periodontitis, the organisms colonizing the root canalspace are growing in verdant biofilms.

One might suspect that the organisms capable of surviving treatment rendered10 years ago and then re-treated to a very high standard with 30 days of Ca(OH)2would show a less verdant biofilm structure, given the extreme nutrient scarcity insuch cases, and that such a biofilm would consist of only a very small number ofspecies. Certainly the culturing model and the literature predict that (21). Thiswork attempts to test that viewpoint by carefully examining one such case in greatdetail.

Materials and MethodsThe patient was a 39-year-old woman in good general health with no signif-

icant medical history. She presented in 2005 with periodic spontaneous pain andtenderness at tooth #30. Clinical tests revealed #30 was mildly tender to percussion.It had a periapical radiolucency associated with the distal root and widening of theperiodontal ligament at the apex of the mesial root. There were no significantprobing depths and no evidence of a sinus tract. The adjacent teeth were non-tender and responded normally to cold. According to the patient, the originalroot canal treatment was performed in about 1995 and had ‘‘bothered her onand off’’ over the years.

After discussing the options of retreatment, surgery, or extraction, the patientmade the decision to re-treat. Access was made through her metal-ceramic crown,and gutta-percha was removed from 3 canals. There was an untreated distobuccal

Failed Molar Retreatment with Secondary Apical Periodontitis 1303

mailto:[email protected]

Figure 1. Timeline of treatment. (A) Preoperative radiograph; original treatment in 1995. (B) Calcium hydroxide placement. (C) Final obturation. (D) Recallradiograph with lesion on mesial root. (E, F) Apical surgery radiographs.


canal. The 4 canals were prepared with a combination of rotary andhand instruments, all to size 45. Sodium hypochlorite (approximately5%) was used for irrigation throughout. At the end of the appointment,17% ethylenediaminetetraacetic acid (EDTA) was used for 1 minute to

Figure 2. (A) Submitted specimen. (B) Decalcified and sectioned. (C) Gutta-pesampled. (E) Isthmus tissue, thick section. (F) Higher magnification of area in (E

1304 Carr et al.

remove the smear layer. Two percent chlorhexidine was used as thefinal irrigant, also for 1 minute. Working lengths were determinedwith a combination of an electronic apex locator (Root ZX; J. Morita,Tokyo, Japan) and paper points. All canals were patent. There were

rcha in mesiobuccal and mesiolingual canal and isthmus. (D) Typical area).


Figure 3. (A) Semi-thin histologic section showing area of interest. (B) TEM of area (A) showing varied biofilm. (C) Higher magnification of small insert areafrom (B) showing multiple phenotypes within a complex biofilm. (D) Higher magnification of larger insert in (B) showing complex biofilm community. Note thatnone of these communities are evident from the histologic semi-thin section, and there is no evidence of any neutrophilic infiltration.


no internal cracks evident under the surgical operating microscope. Atthe end of the appointment, Ca(OH)2 (Ultracal; Ultradent Products, SaltLake City, UT) was injected into all the canals, and a lentulo spiral wasused to ensure it was all the way to working length. A radiograph

Figure 4. (A) Semi-thin area of interest. (B) TEM of biofilm from (A) with no leucanal gutta-percha. (D) TEM of same area (C) demonstrating dentin, pulpal remnalight microscope histologic sections.


confirmed this. Cavit (3 M ESPE, Minneapolis, MN) was used to tempo-rarily close the access opening.

Eleven days later, the patient had a flare-up. Initially she had pain,which lessened as she developed swelling. Clindamycin 300 mg 4 times

kocyte infiltration evident. (C) Random thin section of area near mesiobuccalnt, and complex biofilm. Note that none of these biofilms are evident from the


Figure 5. Variety of bacterial morphotypes. (A) Unusual segmented gram-negative bacterial organism. (B) Dentin, pulp (tissue remnants), and biofilm bacteria inmatrix. (C) Gram-negative bacteria with unusual cellular inclusions. (D) Biofilm bacteria apparently in the process of shedding cell membrane fragments.


a day was prescribed. The swelling resolved after a few days, and sheremained asymptomatic until her obturation appointment about 2weeks later.

At the obturation appointment the Ca(OH)2 was removed, someadditional instrumentation was performed, and the same irrigantswere used plus a final rinse with alcohol to help dry the canals.

Figure 6. Isthmus area pulp tissue (box insert on left). TEM of same specimen onMultiplying gram-negative and gram-positive bacteria.

1306 Carr et al.

Gutta-percha cones (Diadent .06 taper; Diadent Corp, Vancouver, BC,Canada) were fit and condensed by using a warm vertical techniquein combination with Kerr EWT Sealer (Sybron Dental Technologies,Orange, CA). The porcelain margins at the occlusal edge of the accesscavity were etched with 10% hydrofluoric acid (Ultradent), and theaccess opening was restored with composite (BuildIT; Pentron Clinical

the left. (A) Dentin. (B) Pulp. (C) Biofilm bacteria in a thick glycocaylix. (D)


Figure 7. Bacterial profiles indicating active protein synthesis as well as intact cell walls and cell membranes.


Technologies, Wallingford, CT; and Tetric Ceram; Ivoclar Vivadent, Am-herst, NY) and a 4th generation dentin bonding system (Optibond; Syb-ron Dental Technologies).

In 2007, approximately 30 months later, the patient presentedwith spontaneous pain and tenderness at tooth #30. There were nosignificant probings and no swelling or sinus tract. The distal lesionwas greatly reduced in size, but the mesial root now exhibited radiolu-cency. The patient was presented with the options of apical surgery orextraction. She chose surgery. Her husband, who is a physician, hadalready ordered clindamycin for her. A cone beam computed tomog-raphy scan was ordered because of the proximity of the mandibularcanal and mental foramen.

About a week later the surgery was performed. A buccal full-thicknessflap was reflected, and an osteotomy was made through the buccal plate.Because of the nerve proximity the root was resected several millimeterscoronal to the apex, and the root end was removed in one piece and placedimmediately in modified Karnovsky’s fixative and submitted (Fig. 1).

Figure 8. Bacterial blebbing from gram-negative biofilm bacteria. (B) Membrane


A mesial root-end preparation was made with ultrasonics, anda retrofilling of mineral trioxide aggregate (Dentsply International,York, PA) was placed. The flap was closed with 4 gut sutures, whichwere removed after 2 days. The patient has presented twice forfollow-up appointments at 5 and 12 months. She was symptom-free,but healing was incomplete.

Tissue ProcessingImmediately after removal, the apical root tip was fixed by immer-

sion for 1 week in modified Karnovsky’s fixative consisting of 2% para-formaldehyde and 2.5% glutaraldehyde buffered in 0.2 mol/L sodiumphosphate buffer supplemented with sucrose and adjusted for a pHof 7.2 and an osmolarity of 320 mOsm. Thereafter, the root tip was de-calcified for 8 weeks in 0.25 mol/L EDTA and 4% glutaraldehyde sup-plemented with sucrose. The demineralized apex was sectioned into 4horizontal disks, and then each horizontal section was further subdi-vided in half in the coronal-apical direction. The 8 decalcified sections

-enclosed bleb. (C) Higher magnification of bleb.


were then labeled and photographed with a Zeiss surgical microscope(ProErgo, Germany) (Fig. 2).
The decalcified sections were postfixed in 1% OsO4 for 2 hoursfollowed by 0.5% uranyl acetate for 1 hour and then dehydrated inascending concentrations of ethyl alcohol and embedded in Epon812 (Ted Pella, Redding, CA).

Serial 2-mm survey sections along the entire root length andrandom examinations of the isthmus areas in all sections were per-formed. In all, more than 2000 survey sections were examined fromthe apical tip to the most coronal section. Two-micrometer surveysections were taken by using glass knives in an ultra microtome, stainedwith methylene blue, and examined with a Zeiss AxioObserver micro-scope (Zeiss, Oberkochen, Germany). From those survey sections,areas near the main canal in the isthmus area were selected for trans-mission electron microscope (TEM) examination. Five-micrometerpyramids were created with an LKB Pyramitome (Diversified EquipmentCo, Lorton, VA). From those pyramids, 70-nm thin sections were takenand placed on grids, post-stained with uranyl acetate and lead citrate. Inall, 500 thin sections were examined with a Philips 201 (FEI Co, Hills-boro, OR) TEM at 80 kV at multiple magnifications, and selectedsections were photographed by using Kodak 4489 (Eastman Kodak, Ro-chester, NY) film and scanned by using an Epson 750 (Epson America,Inc, Long Beach, CA) scanner at 1200 dpi. Other than brightness andcontrast, no other image processing adjustments have been made tothe scanned files.

ResultsLight Microscope Examination

Histologically, a mesiobuccal and mesiolingual canal filled withgutta-percha was present the entire length of the specimen. A commu-nicating isthmus was also present the entire length. In the sectionsexamined, several accessory or lateral canals were seen. The histologicassessment of the isthmus tissue showed evidence of pulpal remnantswith minimal vascularity and very few neutrophils present. In addition,no obvious bacterial colonies were seen with the light microscope at anylevel (Figs. 3, 4).

TEM ExaminationBecause random thin-sections were taken, we were surprised to

find bacterial biofilm present in every section examined.Of particular note in many of the sections examined was the

complete absence of any intact neutrophils. Only occasional red bloodcells were seen and also very few planktonic organisms. Nearly allbacteria were seen to be encased in a matrix that was many layers thickcomposed of bacteria in various stages of degeneration, with apoptoticcells, ghost cells, and dividing cells all present in the biofilm. In somesections, bacterial biofilm appeared to reside in the dentinal tubules andin interstitial areas of the dentin or fibrodentin. The varied morphotypesseen in this environment differ considerably from those morphotypesobserved in prior studies in which there was not significant nutrientdeprivation, and the morphotypes more closely resembled planktonicforms (Fig. 5 A, C).

Another unsuspected finding was that the appearance of the bio-film differed significantly from one area to the other, even areas invery close proximity. Thus the biofilm community present in this toothwas not a homogeneous, uniform biofilm but one of incredible varietyand changeability. Many different morphotypes were observed in thematrix, and the composition of the biofilm changed dramatically evenbetween adjacent sections 2 mm away (Fig. 6).

Many of the bacteria appeared to be undergoing a lytic process,and others seemed embedded in a collagenous matrix (Fig. 5B, D). Still

1308 Carr et al.

many other areas showed bacterial profiles with many ribosomessuggestive of active protein synthesis and secretion (Fig. 7).

In many areas bacterial blebbing was observed. High magnifica-tion of these blebs suggests these are cell membrane fragments of thebacterial cell membrane (22, 23) (Fig. 8).

DiscussionThere are significant limitations to ultrastructural examinations,

especially with regard to biofilm investigations. Biofilms are collectionsof organisms that exist in a hydrated matrix called extrapolymericsubstance (EPS) that is almost completely lost during tissue processing.TEM and scanning electron microscope imaging both occur in a veryhigh vacuum (10E-10 bar), and all water is lost during specimen prep-aration, so what is seen under the microscope is a very poor substitutefor understanding how these organisms are really structured in thesecommunities (24–29). Because the extracellular milieu in biofilms ismostly water and that water is completely lost in processing, we areleft to really guess how these cells are actually connected and whatcomplexities these communities might have in their fully hydratedand natural state (24–26). Nevertheless, even with these limitations,ultrastructural studies can help elucidate the structure and compositionof endodontic biofilms.

Previous ultrastructural studies have attempted to narrow thefocus of ultrastructural investigation to areas thought likely to containconcentrations of organisms. It is a well-established fact that a lightmicroscope examination of specimens frequently fails to identify organ-isms that are known to be present, even when using special bacterialstains to aid in identification. Identifying bacteria within a biofilm iseven more problematic than identifying colonies of planktonic organ-isms because biofilm bacteria are encased in a polysaccharide matrix.

Previous studies have narrowed the areas investigated to areas inwhich there is an obvious infiltration of neutrophils as seen in the lightmicroscope during the survey sections (19, 20). Our approach wasdifferent because part of the biofilm disease model is that mature bio-films effectively hide themselves from immune surveillance and thatbiofilm organisms might be present where there is no neutrophilicconcentration of organisms (26, 27). For this reason, random sampleswere taken from isthmus areas. In every sample taken, biofilm organ-isms were present, suggesting that limiting areas of investigation to onlywhere neutrophils are observed might result in missing significant bio-film populations.

Another striking finding was the morphologic diversity displayedand how different these organisms appeared compared with prior pub-lished studies (18–20). TEM examination of biofilm bacteria in theirnatural habitat reveals them to be significantly different morphologicallythan their planktonic cousins or even than biofilm bacteria that are inenvironments with greater access to nutrients. The extent and signifi-cance of this morphologic diversity need further investigation. Theauthors want to emphasize that although there is an apparent widemorphologic diversity present in this case, ascribing pathogenicitybecause of this diversity would be an error.

The observation of bacterial blebbing is also of interest. Bacterialblebbing or membrane vesicle formation is a known mechanism thatgrowing gram-negative bacteria use to establish a colonization niche,transmit virulence factors, modulate host defense and response, andeven kill other bacteria. These vesicles can critically impact diseasebecause of their high inflammatory capability. These vesicles typicallycontain lipopolysaccharides of the cell membrane, adhesins, toxins,proteases, alkaline phosphatase, and other immunomodulatorycompounds and are one of the primary ways in which gram-negativebacteria interact with prokaryotic and eukaryotic cells in their environ-ment (22, 30).


In the present study, we saw both gram-positive and gram-negative
organisms surviving in an extremely harsh and nutrient-deficient envi-ronment that had existed for more than a decade. These organismsappear to be in the biofilm mode of growth, which is a consortium ofspecies in a complex matrix in which cell metabolism is low, exopoly-meric substance (EPS) is high, and the arrangement of organisms ishighly variable from region to region. In addition, they do not appearultrastructurally similar to their planktonic cousins, and it appearsthat multiple stressors have induced sublethal and structural changes,resulting in organisms that would probably not be culturable, butthat are still persistent and viable, a hallmark sign of a biofilm. Thesesessile, embedded microbes have drastically down-regulated rates ofcell division, giving them a persistence that enables survival in extremelyharsh environments (23, 27, 31, 32). Factors affecting the survival ofbacteria in this hostile environment include not only nutrient depriva-tion but immune attack by host cells, osmotic stress, chemical insult,oxidative stress, hostile pH changes, desiccation, autochthonous micro-bial toxins, bacteriophage attack, and mechanical physical injury.However, all of these stressors have been shown to be ineffective in elim-inating bacteria within natural biofilms, and many studies have beenpublished that document the incredible persistence of biofilm bacteriadespite such stressors. Once nutrient availability is reestablished, forexample, if the apical or coronal seal allows exposure to serum, theseorganisms can alter their gene expression to accommodate to their newenvironment (23, 31–35).

ConclusionThe biofilm community in this one case appears to have been

incredibly adaptive and persistent, despite a hostile environment thatwould easily kill planktonic organisms. Ultrastructurally they appearfar different from their planktonic brethren, and it will take furtherinvestigations to fully understand all the different forms they mighttake. A fuller understanding of the complex natural life cycle of bacteriawill enable us to gain a better understanding of chronic bacterial infec-tion and its role in endodontic failure. This experiment attempts to showfrom an ultrastructural viewpoint that the microbial communities thatexist within the endodontic space for decades after treatment can thriveand prosper there with no apparent nutrient source and that theircommunities consist of diverse, multispecies biofilms.

References1. Eriksen HM. Epidemiology of apical periodontitis. In: Ørstavik D, Thomas PF, eds.

Essential endodontology: prevention and treatment of apical periodontitis. London:Blackwell Science; 1998:179–91.

2. Kakehashi S, Stanley HR, Fitzgerald RJ. The effects of surgical exposures of dentalpulps in germ-free and conventional laboratory rats. Oral Surg Oral Med Oral Pathol1965;20:340–9.

3. Sundqvist G. Bacteriological studies of necrotic dental pulps. Umea University Odon-tological Dissertations, No. 7, Sweden, 1976.

4. Moller AJR, Fabricius L, Dahlen G, Ohman AE, Heyden G. Influence of periapicaltissues of indigenous oral bacteria and necrotic pulp tissue in monkeys. Scand JDent Res 1981;89:475–84.

5. Aas JA, Paster BJ, Stokes LN, Olsen I, Dewhirst FE. Defining the indigenous micro-biota of the healthy oral cavity. J Dent Res 2003;83 IADR abstract no.26025.

6. Sundqvist G, Figdor D, Persson S, Sjogren U. Microbiologic analysis of teeth withfailed endodontic treatment and the outcome of conservative re-treatment. OralSurg Oral Med Oral Pathol Oral Radiol Endod 1998;85:86–93.


7. Chavez de Paz LE, Molander A, Dahlen G. Gram-positive rods prevailing inteeth with apical periodontitis undergoing root canal treatment. Int Endod J2004;37:579–87.

8. Molander A, Reit C, Dahlen G, Kvist T. Microbiological status of root-filled teeth withapical periodontitis. Int Endod J 1998;31:1–7.

9. Pinheiro ET, Gomes BP, Ferraz CC, Sousa EL, Teixeira FB, Souza-Filho FJ. Microor-ganisms from canals of root-filled teeth with periapical lesions. Int Endod J 2003;36:1–11.

10. Hancock HH III, Sigurdsson A, Trope M, Moiseiwitsch J. Bacteria isolated afterunsuccessful endodontic treatment in a North American population. Oral SurgOral Med Oral Pathol Oral Radiol Endod 2001;91:579–86.

11. Portenier I, Waltimo TMT, Haapasalo M. Enterococcus faecalis-the root canalsurvivor and ‘star’ in post-treatment disease. Endod Topics 2003;6:135–59.

12. Figdor D. Microbial aetiology of endodontic treatment failure and pathogenic prop-erties of selected species. Aust Endod J 2004;30:11–4.

13. Love RM. Enterococcus faecalis-a mechanism for its role in endodontic failure. IntEndod J 2001;34:399–405.

14. Chavez de Paz L. Redefining the persistent infection in root canals: possible role ofbiofilm communities. J Endod 2007;33:652–62.

15. Sunde PT, Olsen I, Debelian GJ, Tronstad L. Microbiota of periapical lesions refrac-tory to endodontic therapy. J Endod 2002;28:304–10.

16. Distel JW, Hatton JF, Gillespie MJ. Biofilm formation in medicated root canals. J En-dod 2002;28:689–93.

17. Svensater G, Bergenholtz G. Biofilms in endodontic infections. Endod Topics 2004;9:27–36.

18. Nair PN, Henry S, Cano V, Vera J. Microbial status of apical root canal system ofhuman mandibular first molars with primary apical periodontitis after ‘‘one-visit’’endodontic treatment. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2005;99:231–52.

19. Nair PNR, Sjogren U, Figdor D, Sundkvist G. Persistent periapical radiolucencies ofroot filled human teeth, failed endodontic treatments and periapical scars. Oral SurgOral Med Oral Pathol 1999;87:617–27.

20. Nair PN, Sjogren U, Krey G, Kahnberg KE, Sundqvist G. Intraradicular bacteria andfungi in root-filled, asymptomatic human teeth with therapy-resistant periapicallesions: a long-term light and electron microscopic follow-up study. J Endod1990;16:580–8.

21. Sundqvist G, Figdor D. Life as an endodontic pathogen: ecological differencesbetween the untreated and root-filled root canals. Endodontic Topics 2003;6:3–28.

22. Li Z, Clarke AJ, Beverage TJ. Gram-negative bacteria produce membrane vesicleswhich are capable of killing other bacteria. J Bacteriol 1998;180:5478–83.

23. Hall-Stoodley L, Costerton JW, Stoodley P. Bacterial biofilms: from the environmentto infectious disease. Nature Reviews Microbiology 2004;2:95–108.

24. Costerton JW. Introduction to biofilm. Int J Antimicrob Agents 1999;11:217–21.25. Stoodley P, Sauer K, Davies DG, Costerton JW. Biofilms as complex differentiated

communities. Annu Rev Microbiol 2002;56:187–209.26. Fux CA, Costerton JW, Stewart PS. Stoodley survival strategies of infectious biofilms.

Trends Microbiol 2005;13:34–40.27. Donlan RM, Costerton JW. Biofilms: survival mechanisms of clinically relevant

microorganisms. Clin. Microbiol Rev 2002;15:167–93.28. Costerton JW, Lewandowski Z. The biofilm lifestyle. Advances in Dental Research

1997;11:192–5.29. Costerton JW, Stewart PS. Battling biofilms. Sci Am 2001;285:74–81.30. Kuehn MJ, Kesty NC. Bacterial outer membrane vesicles and the host-pathogen inter-

action. Genes & Dev 2005;19:2645–55.31. Donlan RM, Costerton JW. Biofilms: survival and mechanisms of clinically relevant

microorganisms. Clin Microbiol Rev 2002;15:167–93.32. Smith JJ, Howington JP, McFeters GA. Survival, physiological response, and recovery

of enteric bacteria exposed to a polar marine environment. Appl Environ Microbiol1994;60:2977–84.

33. Costerton JW, Stewart PS, Greenberg EP. Bacterial biofilms: a common cause ofpersistent infections. Science 1999;284:1318–22.

34. Donlan RM. Biofilm formation: a clinically relevant microbiological process. ClinInfect Dis 2001;33:1387–92.

35. Costerton JW, Veeh R, Shirtliff M, Pasmore M, Post C, Ehrlich G. The application ofbiofilm science to the study and control of chronic bacterial infections. J Clin Invest2003;112:1466–77.


Ultrastructural Examination of Failed Molar …Ultrastructural Examination of Failed Molar...

Documents

Transcript of Ultrastructural Examination of Failed Molar …Ultrastructural Examination of Failed Molar...