MEDICAL EDUCATION

Recovery of Periodontopathogenic Bacteria FromEmbalmed Human Cadavers

NELSON WOOD1AND ROGER B. JOHNSON2*

1Department of Periodontics, School of Dentistry, University of Mississippi Medical Center, Jackson, Mississippi2Department of Anatomy, School of Medicine, University of Mississippi Medical Center, Jackson, Mississippi

There is recent interest in recovery of periodontopathogenic bacteria from arterial andbronchial tissues to identify a link between periodontal and cardiovascular or pulmonarydiseases. This interest could provide a useful clinical correlation exercise for gross anatomy.Our objective was to perform a feasibility study to determine whether these bacteria could berecovered from two sites within eight (4 dentate, 4 edentulous) human embalmed cadaversfrom an anatomical dissection laboratory. Bacterial samples were collected from the rightcoronary artery and the right superior secondary bronchus and assayed for the presence andconcentrations of the DNA of A. actinomycetemcomitans, E. corrodens, C. rectus, P.intermedia, P. gingivalis, B. forsythus, T. denticola, and F. nucleatum. Frequencies werecompared using a Kruskal-Wallis H-test. Correlations between the presence of teeth, bacterialspecies, and site were determined by a Spearman’s rho correlation test. A. actinomycetem-comitans and B. forsythus frequencies were different between the sites in edentulous subjects(P � 0.05); the frequency of B. forsythus was different in dentate and edentulous subjects atthe bronchus site (P � 0.05). Numerous significant correlations were identified betweenstrains of bacteria, site, and presence of teeth. Thus, it is possible for the DNA of periodon-topathogenic bacteria to be recovered from human embalmed cadavers. Collection andidentification of these bacteria from these cadavers could be a useful clinical correlationexercise for dental students in a gross anatomy class. Clin. Anat. 18:64–67, 2005.© 2004 Wiley-Liss, Inc.

Key words: periodontal pathogens; coronary artery; bronchus; human; cadaver

INTRODUCTION

There is recent interest in creation of clinical cor-relation exercises for dental gross anatomy to reinforcethe clinical importance of the dissection experience. Itis often difficult to create clinical correlation exercisesfor dental students for reinforcement of the impor-tance of thorax and abdominal dissections. We havedevised an exercise for study of the systemic spread ofbacteria from the oral cavity to the bronchus andcoronary artery, which has implications in the etiologyof pulmonary diseases and atherosclerosis of the cor-onary arteries.

Recent epidemiological studies have reported anassociation between severe periodontitis, cardiovascu-lar disease (De Stefano et al., 1993; Beck et al., 1996;Scannapieco and Genco, 1999; Armitage, 2000; Sey-mour et al., 2002; Teng et al., 2002), and pulmonaryinfections (Scannapieco and Mylotte, 1996; Scanna-pieco, 1999; Scannapieco and Genco, 1999; Scanna-

pieco and Ho, 2001; Scannapieco et al., 2001; Teng etal., 2002). These diseases are reported to be signifi-cantly correlated with the number of lost teeth (Jo-shipura et al., 1996; Lagervall et al., 2003). Previousstudies suggest that pathogenic periodontal (peri-odontopathogenic) bacteria could be recovered fromnon-oral sites, such as within atheromatous plaquesattached to the wall of coronary arteries and from thesurface of the bronchial epithelium. In live subjects,the spread of these bacteria from the oral cavity to thecoronary arteries is a result of bacteremia (Herzberg

*Correspondence to: Dr. Roger B. Johnson, Department of Anat-omy, School of Medicine, University of Mississippi, 2500 NorthState Street, Jackson, Mississippi 39216-4505.E-mail: rjohnson@sod.umsmed.edu

Received 23 February 2004; Accepted 31 March 2004

Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ca.20041

Clinical Anatomy 18:64–67 (2005)

© 2004 Wiley-Liss, Inc.

and Meyer, 1996), often produced during cleaning ofthe teeth or coincident to minor dental procedures.Oral bacteria probably spread to the lungs in livesubjects by inhalation of aerosolized droplets or byaspiration of oral secretions, especially in medicallycompromised patients (Scannapieco and Ho, 2001).

Several periodontal pathogenic bacteria have beenrecovered from atherosclerotic plaques in surgicalspecimens (Chiu, 1999; Haraszthy et al., 2000;Nguyen et al., 2003). The authors could not identify areport of recovery of periodontopathogenic bacteriafrom embalmed human cadavers. Thus, we designeda study to determine the feasibility for recovery ofthese bacteria from cadavers used in a human anatomycourse for dental students. If successful, this exercisecould be a useful clinical correlate for dental students.

MATERIALS AND METHODS

Our study was approved by the Institutional Re-view Board of the University of Mississippi MedicalCenter. We studied eight embalmed cadavers, whichwere anatomical specimens in a dental gross anatomycourse. The bodies had been embalmed, refrigerated,and then soaked in a solution of formalin and phenolfour months before dissection. All of the bodies wereCaucasian (2 males, 6 females). The mean age atdeath (�SEM) was 68.5 � 3.2 years for the male and82.8 � 3.9 years for the female bodies. Four subjectshad teeth and four were edentulous. Three of thesubjects had died of a respiratory disease; none haddied of a cardiovascular disease.

During dissection of the thorax, bacterial sampleswere obtained from the right lung and the right cor-onary artery. Specifically, specimens were collectedfrom the right superior secondary bronchus at its junc-tion with the right primary bronchus and from theright coronary artery 1 cm distal to its origin. The

arterial endothelium and bronchial mucosa were ex-posed by incising the wall with sharp scissors. Bacte-rial samples were collected by touching the endothe-lial surface of the coronary artery or mucosal surface ofthe secondary bronchus with three sterile paperpoints. Each paper point contacted the tissue surfacefor 10 sec and was immediately placed into a sterilevial. Vials were packaged and sent to a commerciallaboratory for DNA evaluation (DMDx, MicroDentX,Fort Myers, FL). This laboratory could detect boththe presence of eight periodontal pathogens (A. acti-nomycetemcomitans, P. intermedia, P. gingivalis, E. corro-dens, C. rectus, B. forsythus, T. denticola, and F. nuclea-tum) and their concentrations (negative � pathogen is�0.1% of the total or �103 cells; low � pathogen is1–0.9% of the total 103 cells; moderate � pathogen is0.1–9.9% of the total 104 cells; high � pathogen num-ber is �105 cells). Frequencies of the presence of eachbacterial species were calculated and compared usingthe Kruskal-Wallis H-test for frequency differencesand a Spearman rho correlation test using SPSS forWindows v10.1 (SPSS Inc., Chicago, IL).

RESULTS

In the right coronary artery, there was evidence ofa least one species of periodontopathogenic bacteria inall subjects, but there were no significant differencesin the frequencies of the bacteria between edentulousand dentate subjects (Table 1). In the secondary bron-chus, there was a significant difference in the fre-quency of the presence of B. forsythus between eden-tulous and dentate subjects and the presence of A.actinomycetemcomitans between arterial and bronchialsites in edentulous subjects (P � 0.05) (Table 1).

A Spearman correlation test showed significant cor-relations between the presence of C. rectus and P.gingivalis (r � 0.926), E. corrodens and P. intermedia

TABLE 1. Frequency of the Eight Periodontal Pathogens Within the Right Coronary Arteryand the Right Secondary Bronchus Obtained From Embalmed Human Cadaversa

Pathogen

Right coronary artery Right secondary bronchus

Edentulous Dentate Edentulous Dentate

Neg Low Mod High Neg Low Mod High Neg Low Mod High Neg Low Mod High

A. actinomycetem-comitans 0 0 4 0 1 1 1 1 0 0 1 3** 3 0 2 1P. intermedia 3 1 0 0 2 1 1 0 3 1 0 0 2 1 1 0P. gingivalis 0 2 1 1 2 1 0 1 1 0 2 1 2 0 2 0E. corrodens 1 0 3 0 1 1 2 0 3 0 1 0 2 0 2 0C. rectus 2 0 1 1 3 1 0 0 1 3 0 0 2 2 0 0B. forsythus 0 3 1 0 0 1 3 0 0 0 4 0** 1 2 1 0*T. denticola 3 0 1 0 2 1 1 0 2 2 0 0 1 1 1 1F. nucleatum 3 0 0 1 4 0 0 0 4 0 0 0 3 0 1 0

an � 4 for each group. Neg, negative; Mod, moderate.*Significantly different from right secondary bronchus (edentulous): P � 0.05.**Significantly different from right coronary artery (edentulous): P � 0.05.

Recovery of Periodontal Pathogens 65

(r � 0.929) and inverse correlations between P. inter-media and P. gingivalis (r � �0.904), P. intermedia andC. rectus (r � �0.976), and E. corrodens and C. rectus(r � �0.976) (P � 0.001) within the right secondarybronchus of all subjects. Significant inverse correla-tions between the presence of P. gingivalis and P.intermedia (r � �0.706; P � 0.05) were evident at theright coronary artery site in all subjects. There wasalso a significant inverse correlation for the presenceof B. forsythus between the secondary bronchus andright coronary artery sites (r � �0.707; P � 0.05).When adjusted for edentulism, there were significantcorrelations between the presence of F. nucleatum andP. intermedia (r � 0.772, P � 0.05) and P. gingivalis andC. rectus (r � 0.958; P � 0.005) and significant inversecorrelations between F. nucleatum and A. actinomyce-temcomitans (r � �0.862, P � 0.05), C. rectus and P.intermedia (r � �0.909, P � 0.01), E. corrodens and C.rectus (r � �0.968, P � 0.001), P. gingivalis (r ��0.920, P � 0.001) and P. intermedia and P. gingivalis(r � �0.858, P � 0.05) at the right secondary bron-chus site.

DISCUSSION

Our data suggest that periodontopathogenic bacte-rial DNA can be recovered from cardiac and pulmo-nary sites in embalmed cadavers using a laboratorytest that does not require analysis of live bacteria. Toour knowledge, this is the first report of this type ofrecovery and suggests a valid basis for a clinical cor-relate exercise for dental gross anatomy students. Ourmethods have fidelity with those utilizing arterial sur-gical specimens, as similar pathogenic species wererecovered from the surgical samples (Chiu, 1999; Ha-raszthy et al., 2000).

Individual variations in the presence and concen-trations of each pathogen were evident in each sub-ject. This data, taken together with the significantcorrelations of bacterial presence at the small numberof evaluated sites, provide additional evidence to sup-port the hypothesis that the presence of these bacteriaat the bronchus and coronary artery sites is not apost-mortem artifact. Only B. forsythus frequencieswere significantly different between both dentate andedentulous subjects and between right coronary arteryand right secondary bronchus sites. In addition, A.actinomycetemcomitans frequencies were significantlydifferent between right coronary artery and right bron-chus sites in edentulous subjects.

It is possible that the periodontal pathogens recov-ered from the cadavers could have passed from theoral cavity to the bronchus and coronary artery as aresult of the embalming procedure or the soaking of

the body in fixative in the anatomy laboratory. Thesignificant differences in bacterial presence and con-centration between subjects suggests, however, thatsignificant numbers of bacteria were not dislodgedand relocated by the embalming procedure. That sit-uation would have probably resulted in haphazarddistribution of the periodontal pathogens, negatingthe significance of the differences between thegroups. It seems reasonable that these bacteriareached the right coronary artery and the right sec-ondary bronchus during the life of the subject, andwere not post-mortem artifacts. These observationshave clinical implications, as bacterial lipopolysaccha-rides (LPS), not the live bacteria, are reported topromote enhanced release of the superoxide anion(Aida et al., 1995; Skaleric et al., 2000), promotinginflammation. The bacterial LPS could serve as asource of inflammatory stimuli at non-oral sites, whichcould be continuous and long-term. Because peri-odontopathogenic bacterial LPS have variable andspecific effects on neutrophils (Aida et al., 1995) andfibroblasts (Skaleric et al., 2000), their phenotype andconcentration are important information for assess-ment of their potential toxicity.

The presence of periodontopathogenic bacteria inedentulous subjects was unexpected, but supportiveof the observations of others using non-fixed speci-mens. Several periodontopathogenic bacteria havebeen identified within the oral cavity of edentuloussubjects, including P. intermedia (Danser et al., 1995)and P. gingivalis (Danser et al., 1998). A. actinomycetem-comitans, however, is reported to be absent within theoral cavity of edentulous subjects (Danser et al., 1995).Thus, the presence of A. actinomycetemcomitans withinthe coronary arteries and bronchi of edentulous cadav-ers strongly suggests that those bacteria had beendispersed to the non-oral sites when teeth werepresent, and remained there after the extraction of theteeth.

There seems to be merit for using cadavers from agross anatomy laboratory for study of the systemicdistribution of periodontopathogenic bacteria in hu-mans. Although surgical specimens are also useful forthese studies (Haraszthy et al., 2000; Nguyen et al.,2003), human anatomical specimens are readily acces-sible and are available at most medical and dentalschools. Our results suggest that use of anatomicalspecimens in conjunction with a DNA test not requir-ing live bacteria could yield valuable information con-cerning the spread of periodontal pathogens to non-oral sites and could be a viable adjunct model forfuture studies of this phenomenon at other sites af-fected by periodontal bacteremia. Because periodon-tal disease has also been reported to be associated

66 Wood and Johnson

with aortal valve disease (Deshpande et al., 1998),diabetes mellitus (Karjalainen et al., 1994), birth oflow-birthweight infants (Offenbacher et al., 1996),obesity (Wood et al., 2003), osteoporosis (Jeffcoat etal., 2000), and cerebrovascular disease (Wu et al.,2000), additional sites could be evaluated for the pres-ence of periodontopathogenic bacterial DNA, provid-ing an opportunity to offer many types of clinicalcorrelation exercises using anatomical cadavers. Thebacterial collection exercise allows students to learnprocedures for collection and analysis of research dataand could be a useful clinical correlation exercise for adental gross anatomy class. Use of adequate controls(non-periodontopathogenic bacteria), a dental exami-nation of each cadaver, and refinement of our collec-tion techniques could also allow anatomical cadaversto be useful subjects for research of the systemicspread of oral bacteria and their role in the pathogen-esis of a variety of systemic diseases.

ACKNOWLEDGMENTS

The authors thank Dr. N.A. Moore and Dr. D.E.Haines of the Department of Anatomy for their assis-tance and the Mississippi Body Donor Program. Theauthors also thank Dr. C.F. Streckfus for assistance inobtaining analysis of the samples.

REFERENCES

Aida Y, Kukita T, Takada H, Maeda K, Pabst MJ. 1995.Lipopolysaccharides from periodontal pathogens primeneutrophils for enhanced respiratory burst: differential ef-fect of a synthetic lipid as precursor IVA (LA-14-PP). JPeriodont Res 30:116–123.

Armitage GC. 2000. Periodontal infections and cardiovasculardisease—how strong is the association? Oral Dis 6:335–350.

Beck J, Garcia R, Heiss G, Vokonas PS, Offenbacher S. 1996.Periodontal disease and cardiovascular disease. J Periodon-tol 67:1123–1137.

Chiu B. 1999. Multiple infections in carotid atheroscleroticplaques. Am Heart J 138:S534–S536.

Danser MM, Bosch-Tijhof CJ, van Steenbergen TJ, van derVelden U, Loos BG. 1998. Porphyromonas gingivalis in anedentulous proband. A case-report. J Clin Periodontol 25:933–936.

Danser MM, van Winkelhoff AJ, de Graaff J, van der Velden U.1995. Putative periodontal pathogens colonizing oral mu-cous membranes in denture-wearing subjects with a pasthistory of periodontitis. J Clin Periodontol 22:854–859.

De Stefano F, Anda RF, Kahn HS, Williamson DF, RussellCM. 1993. Dental disease and risk of coronary heart diseaseand mortality. BMJ 306:688–691.

Deshpande RG, Khan MB, Genco CA. 1998. Invasion of aorticvalve and heart endothelial cells by Porphyromonas gingivalis.Infect Immun 66:5337–5343.

Haraszthy VI, Zambon JJ, Trevisan M, Zeid M, Genco RJ.2000. Identification of periodontal pathogens in atheroma-tous plaques. J Periodontol 71:1554–1560.

Herzberg MC, Meyer MW. 1996. Effects of oral flora on plate-lets: possible consequences in cardiovascular disease. J Peri-odontol 67:1138–1142.

Jeffcoat MK, Lewis CE, Reddy MS. 2000. Postmenopausalbone loss and its relationship to oral bone loss. Periodontol2000 23:94–102.

Joshipura KJ, Rimm EB, Douglass CW, Trichopoulos D, As-cherio A, Willett WC. 1996. Poor oral health and coronaryheart disease. J Dent Res 75:1631–1636.

Karjalainen KM, Knuuttila MLE, von Dickhoff KJ. 1994. As-sociation of the severity of periodontal disease with organcomplications in type 1 diabetic patients. J Periodontol65:1067–1072.

Lagervall M, Jansson L, Bergstrom J. 2003. Systemic disordersin patients with periodontal disease. J Clin Periodontol30:293–299.

Nguyen S, Bissada N, Weinberg A, De Luca D, McCormick T.2003. Identification of P. gingivalis LPS and IL-1� in ather-omatous plaques of coronary arteries. J Dent Res 82:AADRAbstract #0082.

Offenbacher S, Katz V, Fertik G, Collins J, Boyd D, Maynor G,McKaig R, Beck J. 1996. Periodontal infection as a possiblerisk factor for preterm low birth weight. J Periodontol 67:1103–1113.

Scannapieco FA. 1999. Role of oral bacteria in respiratoryinfection. J Periodontol 70:793–802.

Scannapieco FA, Genco RJ. 1999. Association of periodontalinfections with atherosclerotic and pulmonary diseases. JPeriodont Res 34:340–345.

Scannapieco FA, Ho AW. 2001. Potential associations betweenchronic respiratory disease and periodontal disease: analysisof National and Nutrition Examination Survey III. J Peri-odontol 72:50–56.

Scannapieco FA, Mylotte JM. 1996. Relationships betweenperiodontal disease and bacterial pneumonia. J Periodontol67:1114–1122.

Scannapieco FA, Wang B, Shiau HJ. 2001. Oral bacteria andrespiratory infection: effects on respiratory pathogen adhe-sion and epithelial cell proinflammatory cytokine produc-tion. Ann Periodontol 6:78–86.

Seymour RA, Preshaw PM, Steele JG. 2002. Oral health andheart disease. Prim Dent Care 9:125–131.

Skaleric U, Manthey CM, Mergenhagen SE, Gaspirc B, WahlSM. 2000. Superoxide release and superoxide dismutaseexpression by human gingival fibroblasts. Eur J Oral Sci108:130–135.

Teng YT, Taylor GW, Scannapieco FA, Kinane DF, Curtis M,Beck JD, Kogon S. 2002. Periodontal health and systemicdisorders. J Can Dent Assoc 68:188–192.

Wood N, Johnson RB, Streckfus CF. 2003. Comparison of bodycomposition and periodontal disease using nutritional as-sessment techniques: Third National Health and NutritionExamination Survey (NHANES III). J Clin Periodontol30:321–327.

Wu T, Trevisan M, Genco RJ. 2000. Periodontal disease andrisk of cerebrovascular disease. Arch Intern Med 160:2749–2755.

Recovery of Periodontal Pathogens 67