１

８９

REVIEW

Infective endocarditis and dental procedures : evidence,

pathogenesis, and prevention
Hiro-O Ito

Department of Preventive Dentistry, Institute of Health Biosciences, The University of Tokushima

Graduate School, Tokushima, Japan

Abstract : Infective endocarditis is a serious infection occurring on the endothelial sur-

faces of the heart, especially at the valves. Oral commensal bacteria are the important
etiologic agents in this disease. Common dental procedures, even non-surgical dental
procedures, can often cause bacteremia of oral commensals. Periodontally diseased pa-
tients are at risk from bacteremia even after brushing the teeth. Bacteremia itself rarely
affect healthy people but they can result in mortal infective endocarditis in those who
have a predisposed risk for this disease, such as those with heart valve diseases, pace-
maker implantation, etc. Infective endocarditis is thus established when all the 3 condi-
tions are present simultaneously, i.e., 1) a predisposing impairments in the heart, 2) the
introduction of bacteria into the bloodstream, and 3) the virulence of bacteria. Antibiot-
ics have to be adequately used to prevent this infection, however, their frequent uses
generates drug-resistant mutant bacteria, which is a serious social problem. The
development of novel alternative drugs to be used instead of the current antibiotics is
thus highly desired. We are now using several types of combinatorial peptide libraries
to search for small size molecular mimetics that can interfere with the adhesion of
bacteria to the target organ. The use of such peptides is expected to lead to the develop-
ment of compounds for a novel preventive drug which does not kill bacteria, thus mak-
ing it safer and less likely to generate drug-resistant mutants. J. Med. Invest. 53 : 189-198,
August, 2006

Keywords : infective endocarditis, dental procedures, oral commensal bacteria, viridans streptococci, combinatorial
libraries

from this infection (1), despite improvements in

INTRODUCTION
the outcome due to advances in antimicrobial ther-
Infective endocarditis is a life-threatening dis- apy and an enhanced ability to diagnose and treat
ease. It is always fatal if untreated, and it continues complications. Primary prevention is thus the most
to cause substantial morbidity and mortality de- important. It was first suggested, at the beginning
spite modern antimicrobial and surgical treat- of the twentieth century, that oral bacteria are re-
ments. Substantial morbidity and mortality results lated to infective endocarditis (2, 3). Since then,
interest has grown in the relationship between
Received for publication June 6, 2006 ; accepted July 7, 2006. dental procedures and infective endocarditis, and
Address correspondence and reprint requests to Hiro-O bacteremia as a result of dental treatments has
Ito, Department of Preventive Dentistry, Institute of Health
Biosciences, The University of Tokushima Graduate School,
thus continued to be a subject of research interest.
Kuramoto-cho, Tokushima 770-8504, Japan and Fax : +81-88- Antimicrobial chemotherapy is now widely advo-
633-7338. cated to protect at-risk patients, however, their

The Journal of Medical Investigation Vol. 53 2006

 １
９０ H. Ito Infective endocarditis and dental procedures

frequent uses generates drug-resistant mutant function, and finally results in congestive heart
bacteria, which has become a serious social prob- failure. In addition to the local problems of the
lem (4). heart, pieces of the infective vegetation may break
This short review describes the current knowl- off and travel through the patient’s body through
edge about the relationship between the dental the blood circulation. These infective emboli can
procedures and infective endocarditis and the on- cause cerebral infraction and aneurysms, and thus
going challenge to discover alternative candidate producing infections in remote organs, such as
compounds for use as novel preventive drugs. kidney and spleen. The quick identification of etio-
logic microorganisms and antibiotic therapy based
on the results of microbial sensitivity tests are
INFECTIVE ENDOCARDITIS ; DEFINITION therefore very much desired. However, blood cul-
AND CLASSIFICATION tures often provide negative results and the em-
piric selection of antibiotics is thus necessitated.
Infective endocarditis is an infection of the lining Patients with symptoms of heart failure or with no
of the heart chambers (endocardium), usually the improvement in their clinical conditions despite
heart valves are involved. Fungi, chlamydia, rick- antibiotic treatments therefore have to undergo
ettsia, etc can cause this infection, however, the surgical treatment. The success rate of cardiac sur-
most common causes of this disease are bacteria, gery for infective endocarditis has recently im-
and this phenomenon is therefore also called bac- proved, however, the mortality rate still ranges 20 -
terial endocarditis (5). It has been classified into 25% in Japan (6).
acute and subacute types, according to the natural Men are more susceptible to infective endo-
history of the disease. Recently, as the number of carditis than women ; 1.2 - 3 times more suscepti-
cases with valve replacement surgery is increas- ble for all ages and 2 - 8 times for people older than
ing, the number of endocarditis cases associated 60 years of age (7). The average age of the patients
with the prosthetic valves is increasing. Therefore, with this disease is gradually increasing and it is
these diseases are also classified into “prosthetic now over 55 years of age and about half of all pa-
valve endocarditis” and “native valve endocardi- tients are older than 60 (7). Although the progress
tis”. In addition, the categories of “intravenous of clinical medicine has succeeded in lowering the
drug abuse endocarditis” and “nosocomial endo- mortality of infective endocarditis, the incidence of
carditis” associated with intravenous catheters have this disease has recently been increasing, particu-
been reported (1). larly in developed countries (8). This is reasonable
since owing to the current progress in cardiac sur-
gery, more patients are now surviving who would
PATHOGNENESIS AND TREATMENT not have been rescued previously, this in turn
increases the number of people at high risk from
The establishment of infective endocarditis results infective endocarditis.
from the adherence of microorganisms to wounded
cardiac surfaces and their proliferation at the local
site. Damaged heart valves as a sequela of rheu- ETIOLOGIC AGENTS OF INFECTIVE EN-
matic fever or previous endocarditis, acquired valve DOCARDITIS
lesions, roughened cardiac surfaces as a result of a
jet stream effect from blood crossing congenital The most common causative microorganisms for
cardiac lesions, such as a septal defect, and pros- infective endocarditis are bacteria. Among the wide
thetic heart valves are the usual predisposing clini- variety of bacteria, the leading cause of infective
cal conditions for infective endocarditis. Sterile endocarditis is viridans streptococci (9, 10), par-
vegetations composed of platelet-fibrin clot and ticularly as the cause for the subacute form of this
thrombus are initially formed on the damaged en- diseases. Viridans streptococci comprise the largest
dothelial surfaces, and then, if bacteria are intro- group in the member of streptococci, and they are
duced into the bloodstream, the lesion can trap the most dominant commensals in the oral cavity.
bacterial cells and thus act as an incubator. Subse- The most frequently isolated viridans streptococci
quent growth of the infective vegetation causes from the infective endocarditis patients is S. sanguis
local myocardial abscess that inhibits the valvular (31.9%), followed by S. oralis (29.8), the mutans group
 The Journal of Medical Investigation Vol. 53 August 2006 １
９１

of streptococci, which is notorious for cariogenicity guidelines (5) do not recommend antibiotic pro-
(ability to cause dental caries = tooth cavities), can phylaxis for patients undergoing root canal treat-
also cause infective endocarditis (9). In contrast, ments which do not go beyond the root apex, a
periodontopathic bacteria have been less frequently recent paper (20) reported a marked frequency of
isolated from patients with infective endocarditis. bacteremia following non-surgical root canal treat-
Actinobacillus actinomycetemcomitans has also been ment utilizing several modern reliable techniques
reported to be a causative agent (11), but it is not as to detect bacteria in the blood. Different results can
common as viridans streptococci. Porphyromonas be obtained for the diagnosis of bacteremia by the
gingivalis, the most important pathogen of adult different methodologies for detecting bacterial ex-
periodontitis, has never been isolated from patients istence and the timing of blood sampling.
with infective endocarditis. Regarding the detection methods, there exist a
One should aware of the limitations of the meth- lot of factors that need to be considered, e.g., cul-
odology of bacterial cultures to identify the causes tures in aerobic vs. anaerobic conditions, in broths
of infective endocarditis, although such methods vs. on plates, and compositions of culture media.
have recently greatly improved. When clinical symp- These are the points to be discussed when culture
toms indicate the possibility of infective (bacterial) methods are utilized to detect the presence of bac-
endocarditis, antibiotics may be empirically chosen teria. Recently, a highly sensitive polymerase chain
and given to the patients before sampling their reaction (PCR) technique for detecting bacterial
blood for bacterial cultures, and the antibiotic medi- DNA can be also utilized. The timing of blood
cations in the blood samples can interfere with the sampling is another important consideration for
results of bacterial cultures. In addition, the con- detecting the bacteremia associated with dental
ventional culture conditions are not suited for all procedures, because such bacteremia are known to
kinds of bacteria. For example, nutritionally variant be transient in healthy people who can be enrolled
streptococci, one group of viridans streptococci and in studies for this kind of purpose. Savarrio et al.
recently classified into genera Abiotrophia and (20) have tried to minimize these problems ; they
Granulicatella, have been shown to be causes of so thus obtained samples at three different times re-
-called culture-negative bacterial endocarditis (12). garding the dental procedures, 30 min before the
These factors mentioned here may thus be associ- dental treatment (considered as the negative con-
ated with the low frequency of periodontopathic trol or the background), during the dental proce-
bacteria in infective endocarditis. dures, and 30 min after the end of the dental treat-
ments. Taking these considerations into account,
they have observed bacteremia either during or
INDUCTION OF BACTEREMIA BY DENTAL after non-surgical root canal treatment in 9 out of 30
PROCEDURES patients (33.3%) (20). In their investigation, the most
sensitive method to detect the presence of bacteria
The bloodstream is sterile under normal healthy is shown to be the broth cultures of 15 ml blood
conditions. The introduction of bacteria into the both in aerobic and anaerobic conditions immedi-
bloodstream is necessary for an intracardiac infec- ately after the sampling. The pour plate method using
tion to occur. Such bacteremia occurs after tooth 5 ml blood and PCR detection using 0.2 ml of blood
extractions and other oral surgery procedures. Bac- are shown to be much less sensitive than the broth
teremia as a result of dental treatments has been a culture method. They did not use a lysis filtration
subject of research interest since the first indica- method, in which the host blood cells are lysed and
tion at the start of the 20th century that oral bacteria then the samples are filtrated through membranes
may be related to infective endocarditis (2). Non with 0.2 µm pore size, and bacteria trapped on the
surgical dental procedures have also been reported filter membrane cultured. This method is particu-
to cause bacteremia at significant frequencies. larly useful for blood cultures from patients who are
These include the administration of local anesthe- treated with antibiotics before blood sampling(21).
sia (13), orthodontic band placement (14), peri- The authors of this story(20) did not use the lysis
odontal probing (15), dental prophylaxis (16), scal- filtration method, because the subjects were not
ing and root planing (17), and even after daily tooth given antibiotics; they rather attempted to maximize
brushing (18) and flossing (19). the detection sensitivity by immediate conventional
Although the American Heart Association (AHA) culturing using the two culture techniques. How-
 １
９２ H. Ito Infective endocarditis and dental procedures

ever, they did not try to combine the lysis filtration

method with PCR detection. The combination of
the two methodologies, i.e., “lysis filtration method”
and the “PCR method” allows one to use 15 ml of
blood for PCR detection rather than 0.2 ml used in
their study. This will consequently improve the sen-
sitivity of the PCR method, which may surpass that
of the immediate broth cultures using 15 ml blood.

EXPERIMENTAL INFECTIVE ENDOCARDITIS

Animal models of infective endocarditis have
been established in rabbits (22) and rats (23).
They are useful tools to investigate the effective-
ness of different antibiotics, the pathogenesis of
the disease and bacterial virulence. We previously
investigated the virulence factors of nutritionally
variant streptococci using experimental endocardi-
tis in rats (24) and thus showed their ability to
bind to host extracellular matrix proteins, particu-
larly fibronectin (Figs. 1 and 2). In addition, in an Fig. 1．Comparison of the endocarditis-inducing capacities of
animal model, which demonstrated the same con- nutritionally variant streptococci. The strains derived from in-
ditions as those seen in human infective endo- fective endocarditis patients and oral cavities of healthy volun-
teers were classified by their serotypes. The serotypes I, II and III
carditis, the requirement for the establishment of have a higher capacity to induce endocarditis than IV and V． (29)
the disease are predisposing damage on the endo- ＊Significant difference (P < 0.05) by ANOVA with Tukey’s ad-

justment.
thelial surfaces of the heart and the introduction of
virulent bacterial cells into the bloodstream. Usually
in the animal models, catheters are inserted from
the carotid artery to thus injure the aortic valves
24 hours prior to the intravenous (i.v.) injection of
live bacterial cells. The same i.v. injections of bac-
terial cells to unmanipulated animals produce no
endocardial infection, nor does catheterization
alone. It is also important that all bacteria are not
equally virulent. As shown in Fig. 1, different strains
isolated from oral cavities showed higher or lower
pathogenic capacities, while strains from endocardi-
tis patients regularly showed a high pathogenic
capacity in the rat model. Hence, the three factors,
namely 1) the predisposing impairment on the en-
docardium, 2) the introduction of bacteria in the
bloodstream, and 3) the virulence of bacteria are thus
considered to be necessities for the establishment
of infective endocarditis in both humans and animals
(Fig. 3).
Fig. 2．Extracellular matrix binding of nutritionally variant
streptococci strains. Serotype II (Granulicatella adiacens) and
III strains specifically adhere to fibronectin while serotype I
PREVENTION OF INFECTIVE ENDOCARDI- strains moderately to the ECM proteins tested (Fn: fibronectin,
Lm : laminin, Fbg : fibrinogen). The number of strains included
TIS IN DENTAL PRACTICES in each serotype and their origins are indicated at the bottom
(EC : derived from endocarditis patients, oral : derived from
AHA shows, in 1997, the latest guidelines for the healthy oral cavities) (29)
 The Journal of Medical Investigation Vol. 53 August 2006 １
９３

are performed.

PERIODONTAL DISEASE AND INFECTIVE

ENDOCARDITIS

The problem of whether patients with periodon-

tal diseases are at higher risk from infective endo-
carditis than people who have healthy gingivae
has not yet been fully addressed. However, several
reports indicate that bacteremia is more frequently
inducible in patients with severer periodontal dis-
eases than those who have healthier periodontal
tissue after tooth brushing and periodontal pocket
probing (Table 4) (15, 26). Gum inflammation
loosens the gingival epithelial tissues and it often
Fig. 3．The three conditions are needed to be assembled for becomes ulcerative at the inner part of periodontal
induction of infective endocarditis; 1) predisposing impairment
on the endocardium, 2) introduction of bacteria into the blood- pockets. It can thus provide oral bacteria the route
stream, and 3) virulence of bacteria for getting into the host circulation. Furthermore,
some periodontopathic bacteria such as A. actino-
prevention of infective (bacterial) endocarditis (5). mycetemcomitans and P. gingivalis have the capacity
Patients who are judged to be at risk based on medi- to invade the host epithelial and connective tissues,
cal interviews are thus considered to be appropri- and these characteristics are considered to be in-
ate subjects for antibiotic prophylaxis (Tables 1, 2, volved in the enlargement of inflammatory lesions.
and 3). In Japan, the Japanese Circulation Society Periodontopathic bacteria which seem to be less
has established the guidelines for prevention and important etiologic agents of infective endocarditis
management of infective endocarditis in 2003 (JCS may play important roles in the induction of this
2003) based on the results of a nation-wide hospital disease by causing gum inflammation and opening
questionnaire survey (25). These guidelines are, a route toward the blood circulation for endocardi-
in general, in close accordance with the AHA guide- opathic viridans streptococci. Direct evidence for
lines (5), however, JCS 2003 recommends antibi- the relationship between the prevalence of peri-
otic cover for people using pacemakers and im- odontal disease and the incidence of infective en-
plantable cardioverter defibrillators (ICD), which docarditis remains to be investigated. Cohort studies
are not recommended in the AHA guidelines. Both are expected to provide better evidence to eluci-
guidelines suggest the use of Cephalexin and Ce- date this relationship.
fadoxil for patients with histories of penicillin al-
lergy, but the use of β-lactam agents for patients
with penicillin allergy may be questioned ; other PREVENTION OF INFECTIVE ENDOCARDITIS:
effective agents, such as Clindamycin or Azithro- ALTERNATIVES TO ANTIBIOTICS
mycin/Clarithromycin, should be selected earlier.
Erythromycin has been recommended in the older Antibiotic prophylaxis is the only way to prevent
guidelines (1990), but not in the current one. infective endocarditis, as has already been men-
Regarding the recommendation about dental tioned above. However, the antibacterial spectrum
procedures, the guidelines do not stipulate the ne- is inevitably limited and, moreover, the frequent
cessity of antibiotic prophylaxis for patients who use of antibiotics results in the generation of resis-
are undergoing non-surgical dentistry (Table 2). tant mutant bacteria, which has become a serious
However, as already mentioned above, it has been social problem. Therefore, the development of
shown that bacteremia can be induced by non- safer and hopefully more effective drug alterna-
surgical dental procedures at significant frequen- tives to antibiotics is highly desired. Bactericidal
cies. Therefore, dentists, physicians, and patients or bacteriostatic antibiotics which threaten the life
themselves should be aware of this fact and thus of bacteria induce drug-resistant mutations in order
be as careful as possible when dental treatments to survive. If an agent allows bacteria to live at their
 １
９４ H. Ito Infective endocarditis and dental procedures

Table 1. Cardiac conditions associated with endocarditis (5)

Endocarditis prophylaxis recommended
High-risk category
Prosthetic cardiac valves, including bioprosthetic and homograft valves
Previous bacterial endocarditis
Complex cyanotic congenital heart disease (e.g., single ventricle states, transposition of the great arteries, tetralogy of Fallot)
Surgically constructed systemic pulmonary shunts or conduits
Moderate-risk category
Most other congenital cardiac malformations (other than above and below)
Acquired valvular dysfunction (e.g., rheumatic heart disease)
Hypertrophic cardiomyopathy
Mitral valve prolapse with valvular regurgitation and/or thickened leaflets

Endocarditis prophylaxis not recommended

Negligible-risk category (no greater risk than the general population)
Isolated secundum atrial septal defect
Surgical repair of atrial septal defect, ventricular septal defect, or patent ductus arteriosus (without residua beyond 6 mos)
Previous coronary artery bypass graft surgery
Mitral valve prolapse without valvular regurgitation
Physiologic, functional, or innocent heart murmurs
Previous Kawasaki disease without valvular dysfunction
Previous rheumatic fever without valvular dysfunction
Cardiac pacemakers (intravascular and epicardial) and implanted defibrillators

Table 2. Dental procedures and endocarditis prophylaxis (5)

Endocarditis prophylaxis recommendeda
Dental extractions
Periodontal procedures including surgery, scaling and root planing, probing, and recall maintenance
Dental implant placement and reimplantation of avulsed teeth
Endodontic (root canal) instrumentation or surgery only beyond the apex
Subgingival placement of antibiotic fibers or strips
Initial placement of orthodontic bands but not brackets
Intraligamentary local anesthetic injections
Prophylactic cleaning of teeth or implants where bleeding is anticipated

Endocarditis prophylaxis not recommended

Restorative dentistryb (operative and prosthodontic) with or without a retraction cordc
Local anesthetic injections (non-intra-ligamentary)
Intra-canal endodontic treatment ; post placement and buildup
Placement of rubber dams
Postoperative suture removal
Placement of removable prosthodontic or orthodontic appliances
Taking of oral impressions
Fluoride treatments
Taking of oral radiographs
Orthodontic appliance adjustment
Shedding of primary teeth
a Prophylaxis is recommended for patients with high- and moderate-risk cardiac conditions.
b This includes the restoration of decayed teeth (filling cavities) and the replacement of missing teeth.
c Clinical judgment may indicate antibiotic use in selected circumstances that may create significant bleeding.
 The Journal of Medical Investigation Vol. 53 August 2006 １
９５

Table 3. Prophylactic regimens for dental, oral, respiratory tract, or esophageal proceduresa (5)

Situation Agent Regimen

Standard general prophylaxis Amoxicillin Adults: 2.0 g ; children: 50 mg/kg orally

1 h before procedure
Unable to take oral medications Ampicillin Adults: 2.0 g IM or IV ; children : 50 mg/kg
IM or IV within 30 min before procedure
Allergic to penicillin Clindamycin Adults: 600 mg; children: 20 mg/kg orally
1 h before procedure
Cephalexin or Cefadroxilb Adults : 2.0 g ; children ; 50 mg/kg orally
1 h before procedure
Azithromycin or Clarithromycin Adults: 500 mg; children: 15 mg/kg orally
1 h before procedure
Allergic to penicillin and
unable to take oral medications Clindamycin Adults : 600 mg ; children : 20 mg/kg
IV within 30 min before procedure
Cefazolin Adults : 1.0 g ; children: 25 mg/kg IM or
IV within 30 min before procedure

IM indicates intramuscularly, and IV, intravenously.

aTotal children’s dose should not exceed adult dose.
b Cephalosporins should not be used in individuals with immediate-type hypersensitivity reaction (urticaria, angioedema, or

anaphylaxis) to penicillins.

Table 4. Gingival conditions and induction of bacteremia

Author Nation Measure for Detection Results Statistical

/year /number of periodontal method of relationship
subjects condition bacteremias
(gender)/ages

Sliver Canada Gingival index Blood sampling GI 0∼ 0.75 : 16%

1977a 96 (44 males) (GI) immediately after 2.2∼3.0 : 68% Yes
17∼61 yrs brushing teeth

Daly Australia Alveolar Blood sampling Periodontitis : 40%

2001b 20 periodontitis bone immediately after Gingivitis 10% Yes
(10 males)/43.0 yr level pocket probing
20 gingivitis (radiographs)
(11 males)/35.5 yr
a (26)
b (15)

commensal sites (such as the oral cavities) and can inhibit the bacterial binding to fibronectin have
just inhibit their attachment to a critical site in the been generated. The mAb could be the candidate
body (such as the endocardium), they do not have leading compound of a novel drug to prevent infec-
to mutate for their survival. Such an agent is there- tive endocarditis, however, antibody drugs are not
fore promising to be a novel drug for prevention of feasible for the purpose of preventing infective
infections. The adhesion of bacteria to fibronectin, endocarditis, since it would be very expensive, and
which is a major extracellular matrix protein ex- the repeated use of an antibody drug will result in
posed on the wound endocardium, is considered the generation of a neutralizing antibody by the
to be important for the establishment of this host immune system. We therefore attempted to
infection (24, 27). We have found a novel mode of develop small size molecular mimetics of func-
interaction between endocarditis-causing bacteria tional structures of fibronectin or mAb that can
and human fibronectin (27). During the course of interfere with the bacterial-Fn interaction. For this
this study, monoclonal antibodies (mAb) which purpose, several types of combinatorial peptide li-
 １
９６ H. Ito Infective endocarditis and dental procedures

braries have been constructed ; i.e.,“positional scan- dental procedures, often cause bacteremia that
ning libraries” (Fig. 4) and“on-bead random peptide can result in infective endocarditis in people who
libraries”. Although the real goal of this study, namely have a predisposing risk for this disease, such as
the development of new drugs is still far from valvular heart diseases including prosthetic valves,
being realized, we have nevertheless made some congenital heart diseases, cardiomyopathy, coro-
progress in the methodologies. The findings of some nary artery disease, pacemaker implantation, etc.
of our trials have been described elsewhere (28). The infection is established when all 3 conditions
simultaneously occur, i.e., a predisposing impair-
ment in the heart, the introduction of bacteremia
CONCLUSION and the virulence of the introduced bacteria. Com-
mon dental procedures often cause bacteremia and
Common dental procedures, even non-surgical periodontally diseased patients may even suffer

Fig. 4．The screening of a hexamer positional scanning peptide library by binding of the peptides to an anti-fibronectin mAbs Fynit-
I 01, AB 1-2 (control mAb) and to Fn 105 kDa fragment (negative control protein). The library consists of 114 peptide mixtures (19
amino acids×6 positions, cysteine was omitted from the building block), where each position of the peptide is defined by the amino
acid labeled on the x axis. Each peptide mixture contains 195＝about 2.5×105 different sequences. Individual bars show the levels of
binding of different peptide mixtures at 0.5 mM to the two mAbs and Fn fragment. The six sublibraries are thus expected to provide
information about most important amino acid residues for the molecular recognition at each position (28).
 The Journal of Medical Investigation Vol. 53 August 2006 １
９７

from bacteremia after tooth brushing. Antibiotics 7. Dhawan VK : Infective endocarditis in elderly
have to be used adequately in order to prevent such patients. Clin Infect Dis 34 : 806-812, 2002
infections during dental procedures, however, their 8. Fonager K, Lindberg J, Thulstrup AM, Pedersen L,
frequent use can also generate drug-resistant Schφnheyder HC, Sφrensen HT : Incidence
mutant bacteria. The development of novel drugs and short-term prognosis of infective endo-
to be used as an alternative to the current antibiot- carditis in Denmark, 1980-1997. Scand J Infect
ics is therefore highly desired. Dis 35 : 27-30, 2003
9. Douglas CW, Heath J, Hampton KK, Preston
FE : Identity of viridans streptococci isolated
ACKNOWLEDGMENTS from cases of infective endocarditis. J Med
Microbiol 39 : 179-182, 1993
Our data presented in this paper were obtained 10. Herzberg MC, Meyer MW, Kili! c A, Tao L : Host-
in studies supported in part by Grants-in-Aid for pathogen interactions in bacterial endocarditis :
Scientific Research from the Ministry of Educa- streptococcal virulence in the host. Adv Dent
tion, Science, Sports, Culture and Technology, Japan Res 11 : 69 -74, 1997
(Nos. 11671873, 12557187, 16390540). The author 11. Chen YC, Chang SC, Luh KT, Hsieh WC :
is grateful to Drs. S. Soutome, E. Nagata (Kagoshima Actinobacillus actinomycetemcomitans endo-
University Hospital), Dr. M. Inoue (the Emeritus carditis : a report of four cases and review of
Professor of Kagoshima University), and Drs. K. the literature. Q J Med 81 : 871-878, 1991
Nokihara and T. Ohyama (HiPep Institute Co.) for 12. Ruoff KL : Nutritionally variant streptococci. Clin
their valuable collaboration in these studies. Microbiol Rev 4 : 184-190, 1991
13. Roberts GJ, Simmons NB, Longhurst P, Hewitt
REFERENCES PB : Bacteraemia following local anaesthetic
injections in children. Br Dent J 185 : 295-298,
1. Moreillon P:Endocarditis and endarteritis. In: 1998
Infectious diseases. Armstrong D, Cohen J, eds. 14. McLaughlin JO, Coulter WA, Coffey A, Burden
Harcourt Publishers, London, 1999, pp. 2. 50. 1-2. DJ : The incidence of bacteremia after ortho-
50. 9 dontic banding. Am J Orthod Dentofacial Orthop
2. Horder T : Infective endocarditis. Q J Med 11: 109 : 639-644, 1996
319-323, 1908 15. Daly CG, Mitchell DH, Highfield JE, Grossberg
3. Rushton MA : Subacute bacterial endocarditis DE, Stewart D : Bacteremia due to periodontal
following the extraction of teeth. Guy’s Hosp probing : a clinical and microbiological inves-
Rep 10 : 39-44, 1930 tigation. J Periodontol 72 : 210-214, 2001
4. Mogi A, Nishi JI, Yoshinaga M, Harada H, 16. Fine DH, Korik I, Furgang D, Myers R, Olshan A,
Narahara S, Kawakami K, Maruyama I : In- Barnett ML, Vincent J : Assessing pre-procedural
creased prevalence of penicillin-resistant viri- subgingival irrigation and rinsing with an anti-
dans group streptococci in Japanese children septic mouthrinse to reduce bacteremia. J Am
with upper respiratory infection treated by beta- Dent Assoc 127 : 641-2, 645-6, 1996
lactam agents and in those with oncohema- 17. Bender IB, Naidorf IJ, Garvey GJ : Bacterial en-
tologic diseases. Pediatr Infect Dis J 16 : 1140- docarditis : a consideration for physician and
1144, 1997 dentist. J Am Dent Assoc 109 : 415-420, 1984
5. Dajani AS, Taubert KA, Wilson W, Bolger AF, 18. Sconyers JR, Crawford JJ, Moriarty JD : Rela-
Bayer A, Ferrieri P, Gewitz MH, Shulman ST, tionship of bacteremia to toothbrushing in pa-
Nouri S, Newburger JW, Hutto C, Pallasch tients with periodontitis. J Am Dent Assoc 87 :
TJ, Gage TW, Levison ME, Peter G, Zuccaro 616-622, 1973
G : Prevention of bacterial endocarditis. Rec- 19. Faden HS : Letter: Dental procedures and bac-
ommendations by the American Heart Asso- teremia. Ann Intern Med 81 : 274, 1974
ciation. JAMA 277 : 1794-1801, 1997 20. Savarrio L, Mackenzie D, Riggio M, Saunders
6. Matsushita M, Tsuruya Y, Saitoh M : Myocaditis, WP, Bagg J : Detection of bacteraemias during
pedicarditis, and infective endocarditis, their non-surgicalroot canal treatment. J Dent 33 :
prognosis. Junkankika 51 (in Japanese) : 67-72, 293-303, 2005
2002 21. Heimdahl A, Hall G, Hedberg M, Sandberg
 １
９８ H. Ito Infective endocarditis and dental procedures

H, Soder PO, Tuner K, Nord CE .: Detection and and 2001. Circ J 67 : 901-905, 2003
quantitation by lysis-filtration of bacteremia after 26. Silver JG, Martin AW, McBride BC : Experi-
different oral surgical procedures. J Clin Mi- mental transient bacteraemias in human sub-
crobiol 28 : 2205-2209, 1990 jects with varying degrees of plaque accumu-
22. Baddour LM, Christensen GD, Lowrance JH, lation and gingival inflammation. J Clin Peri-
Simpson WA : Pathogenesis of experimental odontol 4 : 92-99, 1977
endocarditis. Rev Infect Dis 11 : 452-463, 1989 27. Ito H-O, Soutome S, Nokihara K, Inoue M :
23. Kitada K, Inoue M, Kitano M : Experimental Identification and characterization of bacterial
endocarditis induction and platelet aggrega- -binding property in the type III repeat do-
tion by Streptococcus anginosus, Streptococcus main of fibronectin. Biochem Biophys Res
constellatus and Streptococcus intermedius. FEMS Commun 320 : 347-353, 2004
Immunol Med Microbiol 11 : 25-32, 1996 28. Ito H-O, Nokihara K, Soutome S, Ohyama T :
24. Okada Y, Kitada K, Takagaki M, Ito H-O, Inoue Screening of peptides that inhibit bacterial
M : Endocardiac infectivity and binding to ex- binding to fibronectin using combinatorial
tracellular matrix proteins of oral Abiotrophia peptide libraries. Int J Pept Res Ther ( in press,
species. FEMS Immunol Med Microbiol 27 : 2006)
257-261, 2000 29. Okada Y, Ito H-O, Kitada K, Inoue M : Infectivity
25. Nakatani S, Mitsutake K, Hozumi T, Yoshikawa J, of nutritionally variant streptococci : capacity
Akiyama M, Yoshida K, Ishizuka N, Nakamura K, of endocarditis induction and extracellular matrix
Taniguchi Y, Yoshioka K, Kawazoe K, Akaishi M, binding. In : Streptococci and Streptococcal
Niwa K, Nakazawa M, Kitamura S, Miyatake K : diseases. Martin DR, Tagg JR, eds. Secura-
Current characteristics of infective endocardi- copy, Priura, New Zealand, 2000, pp.653-656
tis in Japan : an analysis of 848 cases in 2000