Jan Lindhe. Clinical Periodontology

Подождите немного. Документ загружается.

TRAUMA FROM OCCLUSION • 363

Fig. 15-14. Illustration of a tooth where subgingival plaque has mediated the development of an infiltrated soft tis

sue (shadowed area) and an infrabony pocket (a). When trauma from occlusion of the jiggling type is inflicted (

arrows) on the crown of this tooth (b), the associated pathologic alterations occur within a zone of the

periodontium which is also occupied by the inflammatory cell infiltrate (shadowed area). In this situation the

increasing tooth mobility may also be associated with an enhanced loss of connective tissue attachment and

further downgrowth of dentogingival epithelium; compare arrows in (c) and (d). Occlusal adjustment will result in

a narrowing of the periodontal ligament, less tooth mobility, but no improvement of the attachment level (d) (

Lindhe & Ericsson 1982).

the periodontal structures could adapt, the progressive

increase of the tooth mobility terminated within a few

weeks. The active bone resorption ceased but the an-

gular bone destruction persisted as well as the in-

creased tooth mobility. The periodontal ligament had

an increased width but a normal tissue composition.

Histologic examination of biopsies revealed that this

adaptation had occurred with no greater apical prolif-

eration of the dentogingival epithelium than was

caused by the plaque-associated lesion (Fig. 15-10a,b;

Meitner 1975). This means that occlusal forces which

allow adaptive alterations to develop in the pres-

sure/tension zones of the periodontal ligament will not

aggravate a plaque-associated periodontal disease (

Fig. 15-11).

If, however, the magnitude and direction of the

jiggling forces were such that during the course of the

study (6 months) the tissues in the pressure/tension

zones could not become adapted, the injury in the

zones of co-destruction had a more permanent character.

The periodontal ligament in the pressure/tension zones

displayed for several months signs of inflammation (

vascular proliferation, exudation, thrombosis, reten-

tion of neutrophils and macrophages, collagen de-

struction). Osteoclasts residing on the walls of the

alveolus maintained the bone-resorptive process,

which resulted in a gradual widening of the periodon-

tal ligament in the pressure/tension zones (Fig. 15-12).

As a consequence, the resulting angular bone destruc-

tion was continuous and the mobility of the teeth

remained progressive. The plaque-associated lesion in

the "zone of irritation" and the inflammatory lesion

in the "zone of co-destruction" merged; the den-

togingival epithelium proliferated in an apical direc-

364 • CHAPTER 15

Fig. 15-15. (a) Periodontal conditions around a tooth which has been exposed to trauma from occlusion (of the jig

gling type) for 300 days in combination with plaque-associated experimental periodontitis. (b), Condition of a con

trol tooth from the same dog in which experimental periodontitis but no jiggling trauma had been in operation.

Note the difference between (a) and (b) regarding the degree of bone destruction and loss of connective tissue at

tachment. Note also in (a) the location of the subgingival plaque at the apex of the root. From Ericsson & Lindhe

(

1982).

tion and periodontal disease was aggravated (Figs.

15-

13 & 15-14; Lindhe & Svanberg 1974).

Similar findings were reported from another ex-

periment in the dog (Ericsson & Lindhe 1982) in which

the effect was assessed of a

prolonged

period of jiggling

force application on the rate of progression of plaque-

associated, marginal periodontitis. Thus, in dogs with

continuing periodontal disease, certain teeth were ex-

posed to jiggling forces during a period of 10 months.

Control teeth were not jiggled. Fig. 15-15a illustrates

the marked periodontal tissue breakdown around a

tooth which for several months was exposed to plaque

infection combined with jiggling trauma and Fig. 15-

15b illustrates a control tooth which was exposed to

plaque infection only.

On the other hand, more short-term experiments in

the monkey (Poison & Zander 1983), evaluating the

effect of

trauma from occlusion

on teeth involved in an

ongoing process of periodontitis, failed to support the

findings by Lindhe & Svanberg (1974) and Ericsson &

Lindhe (1982). Poison & Zander (1983) observed that

trauma superimposed on periodontal lesions associ-

ated with angular bony defects (1) caused increased

loss of alveolar bone but (2) failed to produce addi-

tional loss of connective tissue attachment.

CONCLUSIONS

Experiments carried out in humans as well as animals,

have produced convincing evidence that neither uni-

lateral forces nor jiggling forces, applied to teeth with

healthy periodontium, result in pocket formation or in

loss of connective tissue attachment.

Trauma from

occlusion cannot induce periodontal tissue breakdown.

Trauma from occlusion does, however, result in re-

sorption of alveolar bone leading to an increased tooth

mobility which can be of a transient or permanent

character. This bone resorption with resulting in-

creased tooth mobility should be regarded as a physi-

ologic adaptation of the periodontal ligament and

surrounding alveolar bone to the traumatizing forces,

i.e. to altered functional demands.

In teeth with progressive, plaque-associated peri-

odontal disease, trauma from occlusion may, however,

under certain conditions enhance the rate of progres-

sion of the disease, i.e. act as a co-factor in the destruc

tive process. From a clinical point of view, this know-

ledge strengthens the demand for proper treatment of

plaque associated with periodontal disease. This treat-

ment will arrest the destruction of the periodontal

tissues even if the occlusal trauma persists. A treat-

TRAUMA FROM OCCLUSION • 365

ment directed towards the trauma alone, however, i.e.

occlusal adjustment or splinting, may reduce the mo-

bility of the traumatized teeth and result in some

regrowth of bone, but it will not arrest the rate of

further breakdown of the supporting apparatus

caused by plaque. (For a detailed discussion of treat-

ment of teeth exhibiting increased mobility, see Chap-

ter 30).

REFERENCES

Baer, P, Kakehashi, S., Littleton, N.W., White, C.L. &

Lieberman, J.E. (1963). Alveolar bone loss and occlusal wear.

Periodontics 1, 91.

Belting, C.M. & Gupta, O.P. (1961). The influence of psychiatric

disturbances on the severity of periodontal disease. Journal of

Periodontology 32, 219-226.

Box, H.K. (1935). Experimental traumatogenic occlusion in

sheep. Oral Health 25, 9-25.

Burgett, F., Ramfjord, S., Nissle, R., Morrison, E., Charbeneau, T.

& Caffesse, R. (1992). A randomized trial of occlusal adjust-

ment in the treatment of periodontitis patients. Journal of

Clinical Periodontology 19, 381-387.

Ericsson, I. & Lindhe, J. (1977). Lack of effect of trauma from

occlusion on the recurrence of experimental periodontitis.

Journal of Clinical Periodontology 4, 115-127.

Ericsson, I. & Lindhe, J. (1982). The effect of longstanding jig-

gling on experimental marginal periodontitis in the beagle

dog. Journal of Clinical Periodontology 9, 497-503.

Ewen, S.J. & Stahl. S.S. (1962). The response of the periodontium

to chronic gingival irritation and long-term tilting forces in

adult dogs. Oral Surgery, Oral Medicine, Oral Pathology 15,

1426-1433.

Fleszar, T.J., Knowles, J.W., Morrison, E.C., Burgett, F.G., Nissle,

R.R. & Ramfjord, S.P. (1980). Tooth mobility and periodontal

therapy. Journal of Clinical Periodontology 7, 495-505.

Glickman, I. (1965). Clinical significance of trauma from occlu-

sion. Journal of the American Dental Association 70, 607-618.

Glickman, I. (1967). Occlusion and periodontium. Journal of Den-

tal Research 46, Supplement 1, 53.

Glickman, I. & Smulow, J.B. (1962). Alterations in the pathway

of gingival inflammation into the underlying tissues induced

by excessive occlusal forces. Journal of Periodontology 33, 7-13.

Glickman, I. & Smulow, J.B. (1965). Effect of excessive occlusal

forces upon the pathway of gingival inflammation in hu-

mans. Journal of Periodontology 36, 141-147.

Glickman, I. & Smulow, J.B. (1968). Adaptive alteration in the

periodontium of the Rhesus monkey in chronic trauma from

occlusion. Journal of Periodontology 39, 101-105.

Haupl, K. & Psansky, R. (1938). Histologische Untersuchungen

der Wirdungsweise der in der Funktions-Kiefer-Orthopedie

verwendeten Apparate. Deutsche Zahn-, Mund- and Kiefer-

heilikunde 5, 214.

Karolyi, M. (1901). Beobachtungen fiber Pyorrhea alveolaris.

Osterreichisch-Ungarische Viertel Jahresschrift fur Zahnheilkunde

17, 279.

Karring, T., Nyman, S., Thilander, B. & Magnusson, I. (1982).

Bone regeneration in orthodontically produced alveolar bone

dehiscences. Journal of Periodontal Research 17, 309-315.

Lindhe, J. & Ericsson, 1. (1982). The effect of elimination of

jiggling forces on periodontally exposed teeth in the dog.

Journal of Periodontology 53, 562-567.

Lindhe, J. & Svanberg, G. (1974). Influences of trauma from

occlusion on progression of experimental period ontitis in the

Beagle dog. Journal of Clinical Periodontology 1, 3-14.

Lovdahl, A., Schei, O., Waerhaug, J. & Arno, A. (1959). Tooth

mobility and alveolar bone resorption as a function of oc-

clusal stress and oral hygiene. Acta Odontologica Scandinavica

17, 61-77.

Macapanpan, L.C. & Weinmann, J.P. (1954). The influence of

injury to the periodontal membrane on the spread of gingival

inflammation. Journal of Dental Research 33, 263-272.

Manson, J.D. (1976). Bone morphology and bone loss in peri-

odontal disease. Journal of Clinical Periodontology 3, 14-22.

Meitner, S.W. (1975). Co-destructive factors of marginal periodontitis

and repetitive Mechanical injury. Thesis. Rochester, USA: East-

man Dental Center and The University of Rochester, USA.

Muhlemann, H.R. & Herzog, H. (1961). Tooth mobility and

microscopic tissue changes reproduced by experimental oc-

clusal trauma. Helvetica Odontologia Acta 5, 33-39.

Neiderud, A-M., Ericsson, I. & Lindhe, J. (1992). Probing pocket

depth at mobile/nonmobile teeth. Journal of Clinical Periodon-

tology 19, 754-759.

Nyman, S., Lindhe, J. & Ericsson, I. (1978). The effect of progres

sive tooth mobility on destructive periodontitis in the dog.

Journal of Clinical Periodontology 7, 351-360.

Pihlstrom, B.L., Anderson, K.A., Aeppli, D. & Schaffer, E.M. (

1986). Association between signs of trauma from occlusion

and periodontitis. Journal of Periodontology 57, 1-6.

Poison, A. & Zander, H. (1983). Effect of periodontal trauma

upon infrabony pockets. Journal of Periodontology 54, 586-591.

Posselt, U. & Emslie, R.D. (1959). Occlusal disharmonies and

their effect on periodontal diseases. International Dental Jour-

nal 9, 367-381.

Prichard, J.P. (1965). Advanced periodontal disease. Philadelphia:

W.B. Saunders.

Reitan, K. (1951). The initial tissue reaction incident to orthodon

tic tooth movement as related to the influence of function.

Acta Odontologica Scandinavica 10, Supplement 6.

Rosling, B., Nyman, S. & Lindhe, J. (1976). The effect of system

atic plaque control on bone regeneration in infrabony pock-

ets. Journal of Clinical Periodontology 3, 38-53.

Steiner, G.G., Pearson, J.K. & Ainamo, J. (1981). Changes of the

marginal periodontium as a result of labial tooth movement

in monkeys. Journal of Periodontology 56, 314-320.

Stillman, P.R. (1917). The management of pyorrhea. Dental Cos-

mos 59, 405.

Stones, H.H. (1938). An experimental investigation into the as-

sociation of traumatic occlusion with periodontal disease.

Proceedings of the Royal Society of Medicine 31, 479-495.

Svanberg, G. & Lindhe, J. (1973). Experimental tooth hypermo-

bility in the dog. A methodological study. Odontol<gisk Revy

24, 269-282.

Waerhaug, J. (1979). The infrabony pocket and its relationship to

trauma from occlusion and subgingival plaque. Journal of

Periodontology 50, 355-365.

Waerhaug, J. & Hansen, E.R. (1966). Periodontal changes inci-

dent to prolonged occlusal overload in monkeys. Acta Odon-

tologica Scandinavica 24, 91-105.

Wennstrom, J., Lindhe, J., Sinclair, F. & Thilander, B. (1987). Some

periodontal tissue resections to orthodontic tooth movement

in monkeys. Journal of Clinical Periodontology 14, 121-129.

Wentz, F.M., Jarabak, J. & Orban, B. (1958). Experimental occlusal

trauma imitating cuspal interferences. Journal of Periodontol-

ogy 29,117-127.

CHAPTER

Periodontitis as a

Risk for Systemic Disease

RAY C. WILLIAMS AND DAVID PAQUETTE

Early 1900 beliefs

The concept of risk

Understanding the concept of risk

Periodontitis as a risk for:

Coronary heart disease

Pregnancy complications

Diabetic complications

Respiratory infections

Summary

Throughout the history of mankind, there has been the

belief that diseases and maladies which affect the

mouth, such as periodontal disease, can have an effect

on the rest of the body. Over the centuries, writings

from the ancient Egyptians, Hebrews, Assyrians,

Greeks and Romans to name a few, have all noted the

importance of the mouth in overall health and well-

being. Thus, one could say that the concept linking

periodontitis and systemic disease can be traced back

to the beginning of recorded history and medicine

(O'

Reilly & Claffey 2000).

This chapter examines the emerging new evidence

collected since the early 1990s implicating periodontal

infection as a risk factor for several systemic condi-

tions such as coronary heart disease, preterm low

birthweight infants, diabetes and pulmonary disease.

But first, it is helpful for the student of dentistry to

understand the historical perspective under which

this relationship emerged. The concept of "focal infec-

tion" which emerged around 1900 (but died out in the

1940s) has resurfaced and stimulated much debate

and research to elucidate and understand the true

nature of the role of periodontitis as a risk for systemic

disease.

EARLY 1900 BELIEFS THAT ORAL

INFECTION CAUSED SYSTEMIC

DISEASE

At the beginning of the twentieth century, medicine

and dentistry were searching for reasons to explain

why people became afflicted with a wide range of

systemic diseases. Medicine at that time had very little

insight into what caused diseases such as arthritis,

pneumonia and pancreatitis, to name a few. Then,

through the writings and lectures of principally two

individuals, W. D. Miller and William Hunter, the

concept that oral bacteria and infection were the likely

cause of most of a person's systemic illnesses sud-

denly became very popular (O'Reilly & Claffey 2000).

For the next 40 years physicians and dentists would

embrace the idea that infections, especially those

originating in the mouth, caused most of man's suf-

fering and illness. This era, which came to be known

as the "era of focal infection", can be attributed pri-

marily to a microbiologist in Philadelphia, Willoughby

D. Miller and a London physician, William

Hunter (O'

Reilly & Claffey 2000).

Willoughby Miller was a highly regarded Professor

at the University of Pennsylvania School of Dental

Medicine around the turn of the nineteenth century.

Miller had earlier trained in microbiology with Robert

Koch, a scientific pioneer in microbiology and the

father of the modern "germ theory" of disease. While

under the tutelage of Robert Koch, Miller too became

intensely interested in the role of "germs" or bacteria

in causing diseases. Miller returned to the US follow-

ing his trainings convinced that the bacteria residing

in the mouth could cause or be attributed to most

systemic diseases in patients. In a paper published in

1891, entitled "The human mouth as a focus of infec-

tion", Miller argued that the oral flora caused ostitis,

osteomyelitis, septicemia, pyemia, disturbances of the

alimentary tract, noma, dyptheria, tuberculosis,

syphilis and thrush (Miller 1891). Clearly from this

PERIODONTITIS AS A RISK FOR SYSTEMIC DISEASE • 367

one publication, one can appreciate just how exten-

sively the mouth and oral infection were blamed for

causing systemic disease (O'Reilly & Claffey 2000).

While attending one of Miller's lectures at the In-

ternational Congress of Hygiene, William Hunter, a

physician from the London Fever Hospital, noted that

he and Miller were in strong agreement about the

systemic impact of oral infections or oral sepsis.

Shortly after, Hunter was invited to speak at the open

ing of the Strathcona Medical Building at McGill Uni-

versity in Montreal in 1910. In his address to the

audience, he blamed poor dentistry and the resulting

oral sepsis for causing most of mankind's morbidity.

Hunter remarked that the crowns, bridges and partial

dentures he saw in his patients in London were built

on teeth surrounded by a "mass of sepsis". Indeed this

oral sepsis could explain why most individuals devel-

oped chronic diseases (Hunter 1900, O'Reilly & Claf-

fey, 2000). It is likely that Hunter was referring to the

untreated periodontal disease, caries and defective

restorations in his patients at the London Fever Hos-

pital (Fig. 16-1). But whatever Hunter thought he

observed in the mouths of his sick patients, his speech

at McGill University and his subsequent publication

on the role of sepsis and antisepsis in medicine

(

Hunter 1910) ushered in an era of blaming periodon

tal disease, caries and poor oral hygiene as the pri-

mary cause of systemic illness. The term "oral sepsis"

used by Hunter was replaced with the term "focal

infection" in 1911 (Billings 1912). "Focal infection"

implied that there was a nidus of infection somewhere

in the body, such as periodontitis, which via the blood

stream could affect distant sites and organs. Through

out the 1920s and 1930s, dentists and physicians be

lieved that the bacteria on the teeth and the resultant

infectious diseases such as caries, gingivitis and peri-

odontitis that followed, were a "focus of infection"

that

led to a wide variety of systemic problems. It

became

popular during this period to extract teeth as

a means

of ridding the body of oral bacteria and

preventing

and/or treating diseases affecting the

joints, as well

as diseases of the heart, liver, kidneys

and pancreas (

Cecil & Angevine 1938, O'Reilly &

Claffey 2000).

However, by 1940, medicine and dentistry were

realizing that there was much more to explain a pa-

tient's general systemic condition than bacteria in his

or her mouth. Dentists and physicians realized that (1)

extracting a person's teeth did not necessarily make

the person better or make their disease go away, (2)

people with very healthy mouths and no obvious oral

infection developed systemic disease, and (3) people

who had no teeth and thus no apparent oral infection

still developed systemic diseases (Galloway 1931,

Cecil & Angevin 1938).

By 1950 it was apparent to medicine and dentistry

that oral infections such as dental caries, gingivitis and

periodontitis could not explain why individuals de-

veloped a wide range of systemic diseases. Medicine

was, by this time, making great strides in discovering

Fig. 16-1. Poor oral conditions including large plaque

accumulation, caries, gingivitis and periodontitis were

popularly blamed for causing systemic diseases in the

early 1900s.

the true etiologies of many diseases, and dentistry was

making great strides in the prevention as well as the

treatment of caries and periodontal disease. And so

the era of "focal infection" as a primary cause of

systemic diseases finally came to an end (O'Reilly &

Claffey 2000).

Throughout the second half of the twentieth cen-

tury, several researchers and clinicians continued to

ask the question whether oral infection (and inflam-

mation) might in some way contribute to a person's

overall health, but the reasons given were mostly

speculative. Clinicians continued to propose that bac

teria and bacterial products within the periodontal

pocket, which reached the bloodstream from the

mouth, could surely in some way be harmful to the

body as a whole (Thoden van Velzen et al. 1984).

However, it was not until the last decade of the twen-

tieth century that dentistry and medicine again began

to examine the relationship of oral infection as a risk

for systemic disease. The student of dentistry thus

needs to appreciate the intense focus on oral infection

as a "possible" cause of many systemic diseases from

1900-1950, then the era of retreat from the focal infec

tion theories of disease causation from 1950 to around

1989, and now the careful new look at periodontitis as

a possible risk for systemic disease using discreet

scientific levels of evidence.

THE CONCEPT OF RISK

Since around 1989 dentistry has again turned its atten

tion to examining a possible relationship of oral

infec

tion in contributing to an individual's risk for

systemic

disease. The student should note that

dentistry has

come a long way in how we study such a

relationship.

In the first half of the twentieth century,

the evidence

was primarily anecdotal and based on

testimony and statements from dentists and

physicians. There was

no scientific evidence. At

present, however, a group of

highly trained

epidemiologists, basic scientists and clinical scientists

are using modern scientific method-

368 • CHAPTER 16

Fig. 16-2. In a patient having 28 teeth and with ad-

vanced periodontitis of all teeth, the subgingival sur

face area of infection could be roughly similar to the

surface area represented by the two hands.

ology, to carefully and, hopefully, accurately examine

the relationship of periodontitis to systemic disease.

In 1989, Kimmo Matilla and his co-workers in Fin-

land conducted a case-control study on patients who

had experienced an acute myocardial infarction, and

they compared these patients to control subjects se-

lected from the community. A dental examination was

performed on all of the patients and subjects studied

and a dental index computed. The dental index used

by Matilla was the sum of scores from the number of

carious lesions, missing teeth and periapical lesions,

and probing depth measures to indicate periodontitis

and the presence or absence of pericoronitis. Matilla

and his group reported a highly significant association

between poor dental health, as measured by the dental

index, and acute myocardial infarction. The associa-

tion was independent of other risk factors for heart

attack such as age, total cholesterol, high density lipo-

protein (HDL) triglycerides, C peptide, hypertension,

diabetes and smoking (Matilla et al. 1989).

Matilla's provocative findings generated a great

deal

of interest in the scientific community. Medicine and

dentistry were actively studying the concept of "risk"

and were seeking "risk factors" that might help explain

why certain individuals were more suscepti

ble to

disease or why certain individuals responded

better

than others to treatment of disease. Might it be

possible that periodontitis and dental infection did in

fact confer risk on a person for an acute myocardial

infarction? In looking back, it appears that Matilla's

findings came at a critical time in dentistry and the

study of "risk".

Dental scientists first turned their attention to un-

derstanding the "biologic plausibility" of how perio-

dontitis might confer risk for systemic disease. Offen-

bacher, Page, Loesche, Genco, Beck and their col-

leagues, to name a few scientists, noted that periodon-

titis and the anaerobic subgingival infection associ-

ated with periodontitis certainly had the potential, via

the bloodstream, to reach and affect distant sites and

Fig. 16-3. A highly organized biofilm within a periodon-

tal pocket consisting primarily of Gram-negative an-

aerobic bacteria.

organs. Periodontal investigators noted that a patient

who had, for example, 28 teeth with pocket depths of

6-7 mm and bone loss, had a large overall surface area

of infection and inflammation (Waite & Bradley 1965).

In patients with moderate periodontitis, the surface

area could be the size of the palm of the hand or larger

(

Fig. 16-2). In addition, the subgingival bacteria in deep

periodontal pockets exists in a highly organized

biofilm

(Fig. 16-3). Subgingival biofilms would repre

sent a

large and continuing bacterial load and would

be a

constant source of lipopolysaccharide (LPS) as

well as

Gram-negative bacteria to the bloodstream.

Proinflammatory cytokines such as TNF-a, 1L-113,

IFN-y and PGE

can reach high concentrations in the

tissues of the periodontium. The periodontium can

then serve as a renewing reservoir for spillover of

these

mediators into the circulation. Thus, periodontal

pockets containing a Gram-negative biofilm and peri-

odontitis-affected tissues would have the striking po-

tential to flood the circulation with bacteria, bacterial

products such as LPS and inflammatory cytokines, all

of which could reach all parts of the body and possibly

affect distant sites and organs (Loesche 1994, Beck et

al. 1996, 1998, Offenbacher 1996, Page 1998a, Genco et

al. 1999).

Recognizing that it was indeed possible to compre-

hend how periodontitis could contribute to systemic

disease (the "biologic plausibility" of such an idea),

PERIODONTITIS AS A RISK FOR SYSTEMIC DISEASE • 369

the dental profession then turned its attention to ex-

amining existing epidemiological data to see if a rela-

tionship between periodontitis and systemic disease

could be extracted from these existing data sets. Over

the years, several large epidemiological studies had

been conducted on the health status of various popu-

lations. Examples of these studies are the National

Health and Nutrition Examination Surveys I, II and III

(

NHANES I, II, III) and the Veterans Administration

Normative Aging Study. When these epidemiological

studies were conducted there was no actual study of

the relationship of periodontitis to systemic health

and disease. Further, the measures of dental disease

and periodontitis in these studies were diverse. For

example, one study noted periodontitis "present" if a

pocket depth was greater than 5 mm while another

study noted periodontitis was "present" if a pocket

depth was greater than 4 mm. Another study simply

asked the individual whether or not she/he had peri-

odontitis. Thus the clinical measures of periodontitis

in these epidemiological data sets are diverse at best

and almost non-existent at worst. Still, by just exam-

ining these large epidemiology data sets of human

subjects, dental researchers have found compelling

evidence that periodontitis may be a risk for certain

systemic conditions (Loesche 1994, Beck et al. 1996,

1998, Offenbacher 1996, Garcia et al. 2001).

UNDERSTANDING THE CONCEPT

OF RISK

If periodontitis is in fact a risk for systemic disease

then it is important for the student and the dental

profession as a whole to understand how "risk" is

studied and evaluated. We need to understand what

is meant by "risk factor", "risk indicator", "risk

marker" and "background characteristic" (Beck 1994).

In studying risk, the main question is what exactly

is a "risk factor". Risk factor implies that the factor in

question is part of the cause of the disease. Risk factor

is customarily defined as an exposure that increases

the probability that disease will occur. In contrast, a

risk indicator is a suspected risk factor that is correctly

identified through cross-sectional case control study

designs but there are not yet longitudinal cohort study

data available. Risk markers denote those charac-

teristics that have the ability to predict individuals at

high risk for disease, but are not a likely part of the

causal chain. Finally, background characteristics are

those risk markers that may be strongly predictive of

disease, but at present are not changeable (Table 16-1)

(Beck 1994, Offenbacher 1996).

Using the above distinctions, medicine continues to

focus on identifying exposures that may help explain

why certain individuals develop disease. Through

these efforts it is hoped that in identifying risk factors

one can (1) more accurately identify the cause(s) of a

disease, (2) predict which patients are at high risk for

Table 16-1. Definitions of terms commonly used to

study and to describe "risk"

Risk factor

An exposure that increases the probability

that

disease will occur

Risk indicator

A suspected risk factor that is correctly

identified through cross-sectional study designs

but there are not yet longitudinal study data

Risk marker A characteristic or factor that has the ability to

predict individuals at high risk for disease, but is

not likely to be part of the causal chain

Background

A risk marker or predictor that may be strongly

characteristic predictive of disease but at present is not

changeable

disease or disease progression, (3) aid in the diagnosis

of a disease or condition, and (4) provide clues on how

to better prevent the disease or condition (Lilienfeld

1967, Beck 1994).

Beck and co-workers, as well as other epidemiolo-

gists, teach us that it is not easy to determine whether

a suspected risk factor is actually part of the cause of

disease. This can only be done through experimen-

tation, i.e. the randomized controlled trial, and con-

ducting such human experiments is usually neither

ethical nor possible. For example, to establish smok-

ing as causative in heart disease in a clinical trial one

would have to ask one group of subjects to smoke

heavily over several years while control subjects did

not smoke, and then examine the evidence of heart

disease in the two groups. Thus, criteria were establ-

ished some years ago by Sir Bradford Hill and later

refined by Lilienfeld to help epidemiologists decide

whether the evidence from observational studies will

support interpreting a possible "risk factor" as causal.

It is helpful to understand these criteria, because all of

the evidence collected on periodontitis as a risk for

systemic disease may be categorized under these cri-

teria (Table 16-2) (Lilienfeld 1967, Beck 1994, Offen-

bacher 1996, Beck et al. 1998):

1. Consistency of association:

A risk factor is more likely

to be causal if most, if not all, studies dealing with

the relationship produce similar positive results.

This is especially true if the studies involve differ-

ent populations, methods or time periods.

2. Strength of associations:

A valid study is one that is

free from error. The stronger the association, the

less likely the association is entirely due to error

that might distort the results. The strength of asso

ciation is usually quantified with an odds-ratio in

case control studies and a risk ratio in cohort stud

ies.

3. Correct time sequence:

The potential risk factor must

precede the occurrence of the disease. The inability

of cross-sectional studies to determine whether the

factor occurred before the onset of disease is a major

problem in determining causality and thus whether

it is a true risk factor.

370 • CHAPTER 16

Table 16-2. Criteria for considering a possible risk

factor for a disease as important in causing the

disease

Consistency of A factor is more likely to be causal if all studies

associations

dealing

with

the relationship produce similar

results. This is especially true if the studies

involve different populations, methods or time

periods.

Strength of A valid study is one that is free from error. The

associations

stronger the association, the less likely that the

association is entirely due to error that might

distort the results.

Correct time

The potential factor must precede the

sequence

occurrence of the disease.

Specificity of

If a given factor is related to other diseases, its

associations

association with the disease is less likely to be

interpreted as causal. While a specific

association may be more likely to be causal, lack

of specificity cannot justify rejecting causality,

because many diseases have multiple causes.

Degree of

If a factor is of causal importance, then the risk

exposure

of developing the disease should be related to

(dose-response the degree of exposure to the factor.

effect)

Biologic

The association must make sense in light of

plausibility

current knowledge.

Support from Experimental reproduction of the disease should

experimental occur frequently in animals (humans when

evidence

possible) exposed to the risk factor. Randomized

controlled trials testing interventions to prevent

disease occurrence are strong evidence.

4. Specificity of associations:

If a given factor is related

to other diseases, its association with the disease is

less likely to be interpreted as causal. While a spe-

cific association may be more likely to be causal,

lack of specificity cannot justify rejecting causality

because many diseases have multiple causes; a

single factor may produce a number of different

diseases; and a single factor may be a vehicle for

other factors.

5. Degree of exposure:

If a risk factor is of causal impor-

tance, then the risk of developing the disease

should be related to the degree of exposure to the

risk factor. Accordingly, persons with higher levels

of exposure should be at higher risk for the condi-

tion as compared to those persons with lower levels

of exposure.

Biologic

plausibility:

The association of the risk factor

must make sense in light of current knowledge of

the biology or pathogenesis of the disease.

7. Support from experimental evidence:

Experimental re-

production of the disease should occur frequently

in animals exposed to the risk factor. Randomized

controlled human trials which test the effect of

intervention to prevent disease occurrence are also

strong evidence of a causal relationship.

With the above criteria in mind, periodontal investi-

gators began to examine the large epidemiological

data sets which were available to ask the question: is

there a relationship between existence and severity of

periodontal disease and specific systemic diseases?

Much of the research so far using the existing data sets

has examined a relationship between periodontitis

and coronary heart disease. These findings are sum-

marized in the next section.

PERIODONTITIS AS A RISK FOR

CORONARY HEART DISEASE

To date, a number of investigators have reported that

an association between periodontitis and coronary

heart disease/atherosclerosis may exist (Matilla et al.

1995, Beck et al. 1996, 1998, Garcia & Vokonas 1996,

Genco et al. 1997, Loesche et al. 1998, Mendez et al.

1998, Morrison et al. 1999, Valtonen et al. 1999, Wu et

al. 2000a,b).

Atherosclerosis is a progressive disease process in

which large to medium-sized muscular and large elas

tic arteries become occluded with fibro-lipid lesions (

atheromas). Approximately 50% of the deaths in most

industrialized countries are attributable to the compli

cations of atherosclerosis, with coronary thrombosis

and acute myocardial infarction representing about

half of these (Beck et al. 1996). Beck and his colleagues

note that atherosclerosis and periodontitis appear to

have a number of characteristics in common (Beck et

al. 1996). For example, both diseases are more likely to

occur in persons who are older, male, of lower educa-

tional status, who smoke, are hypertensive, who are

stressed and are socially isolated. Beck and Offen-

bacher suggest that these common characteristics may

indicate that periodontal disease and heart disease

share a similar etiologic pathway (Beck 1994, Beck et

al. 1998, Beck & Offenbacher 1998).

One of the initial compelling reasons for consider-

ing periodontitis as a risk factor for cardiovascular

disease relates to the strong body of emerging evi-

dence on infections or inflammatory etiologies for

atherosclerosis and coronary heart disease. A sum-

mary of the evidence is as follows (Beck et al. 1998):

•

Prior flu-like symptoms are more common in cases

of myocardial infarction than in controls studied at

the same time.

•

High levels of cytomegalovirus antibodies were as-

sociated with elevated carotid intimal-medial wall

thickness 18 years later.

•

Prior infection with cytomegalovirus is a strong

independent risk factor for restenosis after coronary

atherectomy.

•

Dental infections are more common in cases of

cere

bral infarction compared to control subjects

matched for age and gender.

PERIODONTITIS AS A RISK FOR SYSTEMIC DISEASE • 371

Table 16-3. Strength and consistency of associations between oral conditions and atherosclerosis/coronary

heart disease (CHD). From Beck et al. 1998 with permission.

Study

Design

Association

Measure

Mattila et al. 1989/Finland

Case-control

Total dental index and heart attack

OR = 1.

Mattila et al. 1993/Finland

Case-control Total dental index and atheromatosis

OR = 1.

Mattila et al.

1995/Finland

Follow-up

Total dental index and new events

HR = 1.

DeStefano et al. 1993/USA

Cohort

PI, OHI, and admits and death due to CHD

RR = 1.

Beck et al. 1996/Boston

Cohort PD, bone levels, and new CHD, fatal CHD & stroke

OR = 1.5, 1.9,

2.8

Genco et al. 1997/USA

Cohort

Bone loss and new CHD

OR = 2.

Joshipura et al. 1996/USA

Cohort

Tooth loss in men with periodontal disease and CHD

RR = 1.

PI = plaque index; OHI = oral hygiene instruction; PD = probing depth.

Table 16-4. Specificity of associations between oral conditions and atherosclerosis/coronary heart disease

(

CHD). From Beck et al. 1998 with permission.

Study Association

Measure

Adjusted

Mattila et al. 1989/Finland

Total dental index and heart attack

OR = 1.

Yes

Mattila et al. 1993/Finland

Total dental index and atheromatosis

OR = 1.

Yes

Mattila et al. 1995/Finland

Total dental index and new events

HR = 1.

Yes

DeStefano et al. 1993/USA

PI, OHI, and admits and death due to CHD

RR = 1.

Yes

Beck et al. 1996/Boston PD, bone levels, and new CHD, fatal CHD and stroke

OR = 1.5, 1.9,

2.8

Yes

Genco et al. 1997/USA

Bone loss and new CHD

OR = 2.

Yes

Joshipura et al. 1996/USA Tooth loss in men with periodontal disease and CHD

RR = 1.

Yes

Table 16-5. Time sequence for associations between oral conditions and atherosclerosis/coronary heart dis-

ease (CHD). From Beck et al. 1998 with permission.

Study Association

Sequence

Mattila et al. 1995/Finland

Total dental index and new events

Oral condition prior to 7-year

follow-up

of patients

with myocardial infarction

DeStefano et al. 1993/USA

PI, OHI, and admits and death due to CHD

Oral status prior to 14-year

follow-up;

assumed to

be w/o CHD at baseline

Beck et al. 1996/Boston PD, bone levels, and new CHD, fatal CHD and

PD and bone loss prior to 18-year

follow-up

stroke

men w/o CHD

Genco et al. 1997/USA

Bone loss and new CHD

Bone loss prior to 10-year

follow-up

Joshipura et al. 1996/USA

Tooth loss in men with periodontal disease and

Men reporting signs and symptoms of CHD

CHD

excluded from follow-up analysis

•

The gingival index is significantly correlated with

fibrinogen and white cell counts in periodontal pa-

tients and controls, adjusted for age, smoking and

socio-economic status.

•

Plasma levels of cholesterol, low density lipoprotein

cholesterol and triglycerides were significantly

higher in 46 periodontitis patients aged 50-60 com-

pared to age and gender-matched controls. In addi-

tion, successful periodontal therapy was more

likely

in persons with lower lipid values (Beck et al. 1998).

As noted earlier, in order to evaluate the data con-

tained in large existing data sets which may link peri-

odontitis to systemic disease, investigators used the

criteria for studying risk outlined by Lilienfeld (1967).

What follows is a summary of evidence gathered to

date from these epidemiological studies and the data

are presented in Tables 16-3, 16-4 and 16-5.

372 • CHAPTER

Consistency, strength and specificity of

associations

Matilla and co-workers conducted two separate case-

control studies totaling 100 patients with acute myo-

cardial infarction and 102 controls selected from the

community. In Matilla's original 1989 report, subjects

with evidence of oral infection were 30% more likely

to present with myocardial infarction relative to sub-

jects without oral infections (Matilla et al. 1989, Matilla

1989).

In a second case control report, Matilla and co-

workers noted associations between dental infections

and the degree of atherosclerosis. This study exam-

ined the same subjects as the first report but pertained

to information obtained with diagnostic coronary

angiography. Accordingly, the left main coronary ar-

tery, the a, b, and c portions of the right coronary artery,

the circumflex artery, and the left anterior descending

artery were assessed diagnostically and graded for a

degree of occlusion on a 5-point scale. Again, the total

dental index score was used as a general score for

dental caries, endodontics and periodontal infections.

In a multivariate analysis, significant associations

were found between dental infections, age and

triglycerides and severe coronary atheromatosis.

These associations remained significant even after ad

justing for other known risk factors like total choles-

terol, HDL, smoking, hypertension, socio-economic

status and body mass index. The authors postulated

that bacterial infections have profound effects on en-

dothelial cells, monocyte-macrophages, thrombo-

cytes and blood coagulation and lipid metabolism,

and concluded that dental infections are the only fac-

tor outside the scope of the classic coronary risk fac-

tors which have shown an independent association

with the severity of adult coronary atherosclerosis in

their multivariate assessment (Matilla et al. 1993, Beck

et al. 1998). Continuing to monitor for myocardial

infarction among the cases in these first case control

reports, Matilla et al. (1995) presented Cox propor-

tional hazard models further implicating dental infec-

tion as a significant risk factor for new cardiovascular

events.

DeStefano and co-workers assessed the association

between periodontal disease and coronary heart dis-

ease within the National Health and Nutrition Exami-

nation Survey I (NHANES I), which followed subjects

for 14 years. This cohort study examined several po-

tentially confounding variables including age, gender,

race, education, marital status, systolic blood pres-

sure, total cholesterol levels, body mass index, diabe-

tes, physical activity, alcohol consumption, poverty

and cigarette smoking. These investigators reported

that among the 9760 subjects examined longitudinally,

those with periodontitis had a 25% increased risk of

coronary heart disease relative to those with minimal

periodontal disease, adjusted for the co-variables

mentioned above (DeStefano et al. 1993). Interestingly,

males younger than 50 years old with periodontitis

were 72% more likely to develop coronary heart dis-

ease compared to their periodontally healthy counter-

parts.

Using data obtained in the Normative Aging Study,

Beck and co-workers evaluated 921 men aged 21-80

years who were free of coronary heart disease at base

line. Over an 18-year follow-up period, 207 men de-

veloped coronary heart disease, 59 died of coronary

heart disease and 40 had strokes. Odds ratios adjusted

for age and established cardiovascular risk factors

were 1.5,1.9 and 2.8 for periodontal bone loss and total

coronary heart disease, fatal coronary heart disease

and stroke respectively (Bell et al. 1966, Kapur et al.

1972, Beck et al. 1996, Garcia & Vokonas 1996). These

data indicated that persons with radiographic evi-

dence of periodontitis were 0.5 to 2.8 times more likely

to develop coronary heart disease or suffer from a

vascular event.

In a larger six-year cohort study, Joshipura and

co-workers followed 44 119 men in the health profes-

sions who reported no coronary heart disease symp-

toms at baseline, no notable dietary histories and no

missing information on age or number of teeth. All

information was obtained via mailed questionnaires.

Joshipura and her group found that the association

between self-reported history of periodontal disease

and incidence of heart disease was not significant after

adjusting for other risk factors such as age, body mass,

exercise, smoking habits, alcohol consumption, family

history of myocardial infarction before age 60 and

intake of vitamin E (relative risk = 1.04). It should be

noted that most of the studies showing a positive

association between periodontal disease and coronary

heart disease have highlighted the importance of the

extent of periodontal disease burden in the relation-

ship. Since the subjects in this study dichotomously

responded to a question about periodontal disease (

yes/no) it is not possible to quantify the extent of the

periodontal disease present in these participants. In

addition, misclassification from self-reported peri-

odontal disease seems very likely. The study did how-

ever demonstrate that men with tooth loss and peri-

odontal disease were 70% more likely to exhibit coro-

nary heart disease (Joshipura et al. 1996).

Genco and co-workers investigated the association

between periodontal infection and risk of cardiovas-

cular disease in 1372 Native Americans of the Gila

River Indian Community, a group with a high preva-

lence of diabetes mellitus. At baseline, alveolar bone

level was measured and cardiovascular status was

monitored for up to 10 years. Among all age groups,

alveolar bone level was predictive of coronary heart

disease, but did not remain significant in a multivari-

ate analysis. In contrast, for persons younger than 60

years of age, alveolar bone level was predictive of

coronary heart disease, with an odds ratio of 2.68

(

Genco et al. 1997).

A recent report of another examination of the

NHANES I data did not find an association between

periodontal disease and subsequent heart disease.