tongue appears smooth in pernicious anemia. Group B vitamin deficiency is associated with oral mucositis and ulcers, glossitis, and burning sensations of the tongue. Scurvy, caused by severe vitamin C deficien- cy, is associated with gingival swelling, bleeding, ulceration, and tooth loosening. Lack of vitamin D in utero or infancy impairs tooth development. Enamel hypoplasia may result from high levels of fluoride or from disturbances in calcium and phosphate metab- olism, which can occur in hypoparathyroidism, gas- troenteritis, and celiac disease. The mouth also can reflect the effects of tobacco use, perhaps providing the only visible evidence of its adverse effects. Oral Manifestations of HIV Infection and of Osteoporosis The mouth can serve as an early warning system, diagnostic of systemic infectious disease and predic- tive of its progression, such as with HIV infection. In the case where oral cells and tissues have counter- parts in other parts of the body, oral changes may indicate a common pathological process. During rou- tine oral examinations and perhaps in future screen- ing tests, radiographic or magnetic resonance imag- ing of oral bone may be diagnostic of early osteo- porotic changes in the skeleton. The following sec- tions provide details. HIV Infection The progressive destruction of the body's immune system by HIV leads to a number of oral lesions, such as oral candidiasis and oral hairy leukoplakia, that have been used not only in diagnosis but also in determining specific stages of HIV infection (CDC 1992, 1994, Montaner et al. 1992, Redfield et al. 1986, Royce et al. 1991, Seage et al. 1997). Oral can- didiasis is rarely seen in previously healthy young adults who have not received prior medical therapy such as cancer chemotherapy or treatment with other immunosuppressive drugs (Klein et al. 1984). It was associated with AIDS as early as 1981 in the first report of the syndrome (CDC 1981a) and was fre- quently noted among otherwise asymptomatic HIV- positive populations (Duffy et al. 1992, Feigal et al. 1991). Oral candidiasis may be the first sign of HIV infection and often occurs as part of the initial phase of infection—the acute HIV syndrome (Tindall et al. 1995). It tends to increase in prevalence with pro- gression of HIV infection when CD4 lymphocyte counts fall (Glick et al. 1994a, Lifson et al. 1994). It also appears to be the most common oral manifesta- tion in pediatric HIV infection (Kline 1996, Leggott Linkages with General Health 1995, Ramos-Gomez et al. 1996) and has been demonstrated to proceed to esophageal candidiasis, a sign of overt AIDS (Saah et al. 1992). Both the pseudo- membranous and the erythematous forms of candidi- asis appear to be important predictors of progression of HIV infection (Dodd et al. 1991, Klein et al. 1984, 1992). Like oral candidiasis, oral hairy leukoplakia in HIV-positive persons heralds more rapid progression to AIDS (Glick et al. 1994a, Greenspan et al. 1987, Lifson et al. 1994, Morfeldt-Manson et al. 1989). Oral hairy leukoplakia is an oral lesion first reported in the early days of the AIDS epidemic (Greenspan Det al. 1984, Greenspan JS et al. 1985). Since its discov- ery, hairy leukoplakia has been found in HIV-negative persons with other forms of immunosuppression, such as organ or bone marrow recipients and those on long-term steroid therapy (Epstein et al. 1988, Greenspan et al. 1989, Itin et al. 1988, King et al. 1994, Zakrzewska et al. 1995), and less frequently among immunocompetent persons (Eisenberg et al. 1992, Felix et al. 1992). In a comprehensive review of periodontal condi- tions, Mealey (1996) noted that linear gingival ery- thema and necrotizing ulcerative periodontitis may be predictive of progression of HIV infection. Necrotizing ulcerative periodontitis, a more serious periodontal condition observed in HIV-infected per- sons, is a good predictor of CD4+ cell counts of under 200 per cubic millimeter, and in one study was a strong predictor of rapid progression to death (Glick et al. 1994a,b, Winkler and Robertson 1992, Winkler et al. 1988). In addition, the numerous ulcerative and nonulcerative conditions that affect the oral cavity (Aldous and Aldous 1991, Coates et al. 1996, Cruz et al. 1996, Gandolfo et al. 1991, Itin et al. 1993, Mealey 1996) may affect the biologic activity of HIV and are affected by its treatments. Other oral conditions, unexpected in the oral cavity, have been noted in the early stages of HIV infection. The increased incidence of Kaposis sarco- ma among young men in New York and California was one of the earliest signs of the AIDS epidemic (CDC 1981b). In addition, some conditions create a problem for differential diagnosis. For example, because involvement of the gingiva is common, early non-Hodgkin's lymphoma lesions are frequently mis- taken for common periodontal or dental infections (Epstein and Silverman 1992). The changing face of the HIV epidemic and changes in the therapies used to manage complica- tions are reflected in changes in the oral manifesta- tions, which warrant continued surveillance and research. The increasing resistance of microorganisms ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL 101 Linkages with General Health to antibiotics and antifungals is challenging. On the other hand, the completion of the Candida albicans genome may yield better treatments for this oppor- tunistic infection. Osteoporosis and Oral Bone Loss With growing numbers of Americans living longer, there have been concomitant increases in the num- bers affected by age-related chronic degenerative dis- eases. Prominent among these conditions are bone and joint diseases. It is likely, for example, that some temporomandibular joint disorders are manifesta- tions of osteoarthritis, rheumatoid arthritis, or myofascial pain. Paget's disease, characterized by enlarged and deformed bone, can be particularly painful and debilitating when it affects the cranial and jaw bones. Osteoporosis, a degenerative disease character- ized by the loss of bone mineral and associated struc- tural changes, has long been suspected as a risk fac- tor for oral bone loss. In addition, measures of oral bone loss have been proposed as potential screening tests for osteoporosis (Jeffcoat 1998). Osteoporosis affects over 20 million people in the United States, most of whom are women, and results in nearly 2 million fractures per year (National Institute of Arthritis, Musculoskeletal and Skin Diseases 2000). The disease is more prevalent in white and Asian American women than in black women. Oral bone loss has been reported to be more prevalent in women than in men. Studies by Ortman et al. (1989) found a higher percentage of women than men with severe alveolar ridge resorption. This finding parallels the findings of Humphries et al. (1989), who showed that loss of bone mineral densi- ty with age in edentulous adult mandibles was more significant in women than in men. Also, the associa- tion between estrogen status, alveolar bone density, and history of periodontitis in postmenopausal women has been studied (Payne et al. 1997). Most of the studies in this area have examined bone loss in women, and most investigators have reported a correlation between oral and skeletal bone loss measured in a variety of ways. Studies of non- osteoporotic women by Kribbs et al. (1990) showed that mandibular bone mass is significantly correlated with skeletal bone mass. Dual photon absorptiometry measurements of jawbone volume in women with osteoporosis have shown that reduction in mandibu- lar bone mass is directly related to the reduction in total skeletal mass density (Kribbs and Chesnut 1984, Kribbs et.al. 1983, von Wowern 1985, 1988, von Wowern et al. 1994). Kribbs et al. (1989) further showed that mandibular mass is correlated with all skeletal measures in osteoporotic women and that the height of the edentulous ridge is correlated with total body calcium and mandibular bone mineral density. Hirai et al. (1993) found that the presence of skeletal osteoporosis strongly affects the reduction of the residual ridge in edentulous patients. A small case-control study comparing older female patients with osteoporotic fractures and non-osteoporotic women without fractures found greater periodontal attachment loss in the osteoporotic women than in the controls (von Wowern et al. 1994). Studies that have controlled for confounding fac- tors also have found correlations between oral bone loss and skeletal bone density. Controlling for pack- years of smoking, education, body anass, and years since menopause, Krall et al. (1994, 1996) found a significant positive relationship between number of teeth and bone mineral density of the spine and the radius. In a cohort of 70 postmenopausal women, Wactawski-Wende et al. (1996) measured skeletal bone mineral density at the Ward's triangle area of the femur and compared it with periodontal disease assessed by attachment loss and the height of alveo- lar bone measured by radiographs. After adjusting for age, years since menopause, estrogen use, body mass index, and smoking, the investigators concluded that osteopenia (low bone mass) is related to alveolar cres- tal height and tooth loss in postmenopausal women. Methods used to measure oral and skeletal bone loss have varied among investigators and have shown different outcomes. Kribbs (1990) found that patients in an osteoporotic group had lost more teeth, had less mandibular bone, and had a thinner bone measured at a part of the jaw (cortex at the gonion) than a comparable non-osteoporotic group. However, using periodontal attachment loss as an indicator of mandibular bone loss, they found no differences between the osteoporotic and the non-osteoporotic group. Mohajery and Brooks (1992) compared non- osteoporotic postmenopausal women with women with mild to moderate osteoporosis and found no correlation between mandibular and skeletal bone mineral density. This study raises questions about the quantification of mild and moderate osteoporosis. Defining healthy periodontal tissues as having no periodontal pockets deeper than 5 millimeters, Hildebolt et al. (1997) studied postmenopausal women with healthy periodontal tissues and found no relationship between periodontal attachment loss and postcranial bone mineral density. How- ever, preliminary studies from the oral ancillary study of the NIH Women’s Health Initiative report signi- ficant correlations between mandibular basal bone mineral density and hip bone mineral density 102 ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL (r = 0.74, P < .001) Jeffcoat et al. in press). In this study, digital subtraction radiography methods were used for mandibular bone measurements, and dual- energy X-ray absorptiometry (DXA) scans were used for the hip bone measurements. The authors of this study propose the possibility that high-quality intra- oral radiographs may be used in the future for screen- ing osteopenia. Larger cross-sectional studies, as well as longitu- dinal and mechanism studies, are needed to better define the relationship between osteoporosis. osteopenia, and oral bone loss, periodontal disease, and tooth loss. The role of factors involved in the reg- TABLE 5.2 Saliva/oral fluids: sampled analytes and current FDA-approved tests Category Analytes Alcohol Amphetamines Barbiturates Benzodiazepines Methamphetamine Cocaine Opiates LSD PCP Marijuana Ethanol Nicotine Opiates PCP Antibodies? HIV HPV HHV-8/KSH C. parvum Helicobacter pylori FDA-Approved Tests Drugs of abuse? Cannabinoids Cocaine Cotinine HIV antibodies Hormanes¢ Cortisol Estriol Progesterone Testosterone Substance P Met-enkephalin Cadmium Lead Mercury Environmental toxins! Therapeutics? Antipyrine Carbamazepine Ciprofloxacin Irinotecan Lithium Methotrexate Phenytoin Phenobarbital Theophylline aCone et al. 1993, 1997. bConstantine et al. 1997. Dabbs 1993, Ellison 1993. éGonzalez et al. 1997, Joselow et al. 1968. eWilson 1993. Source: Constantine et al. 1997. Linkages with General Health ulation of bone mineral density in men as well as in postmenopausal women needs to be evaluated fur- ther with reference to oral bone loss, tooth loss, and periodontal disease. Variables such as sex, race, dietary calcium and phosphorus, vitamin D intake, exercise, body mass index, smoking, genetics, med- ication use, reproductive history, and psychosocial factors need to be assessed in depth. In addition, reli- able and valid criteria and imaging technologies for assessing osteoporosis and oral bone loss are needed to better elucidate the full relationship between skeletal and mandibular bone mineral density, peri- odontal disease, alveolar ridge resorption, and tooth loss. Oral-fluid-based Diagnostics: The Example of Saliva The diagnostic value of salivary secretions to detect systemic diseases has long been recognized (Mandel 1990), and oral fluids and tissues (buccal cells) are increasingly being used to diagnose a wide range of conditions. Saliva- and oral-based diagnostics use readily available samples and do not require invasive procedures. Researchers have detected antibodies in saliva that are directed against viral pathogens such as human immunodeficiency virus (Malamud 1997) and hepatitis A virus (O'Farrell et al. 1997) or B virus (Richards et al. 1996). Saliva is being used to detect antibodies, drugs, hormones, and environmental tox- ins (Malamud and Tabak 1993) (Table 5.2). The sim- plest tests are those that detect the presence or absence of a substance in the saliva, such as various drugs. Greater technical challenges are presented for tests that will be used for therapeutic monitoring since accurate levels of a substance and/or its metabolites are needed. In these instances the sali- va/plasma concentration ratio must be determined experimentally (Haeckel 1993). Tests beyond those listed in Table 5.2 are currently on the market, but do not yet have FDA approval. Saliva is also the fluid of choice to assess the integrity of the mucosal immune system (Mandel 1990). Most recently, oral fluids have been used as a source of microbial or host DNA. With the advent of polymerase chain reaction methods, the DNA con- tained within a single cell is sufficient for detection of viruses (e.g., Kaposi's sarcoma-associated herpes virus, Koelle et al. 1997; Epstein-Barr virus, Falk et al. 1997, mumps virus, Afzal et al. 1997) or bacteria (e.g., Helicobacter pylori, Reilly et al. 1997). Similarly, DNA extracted from sloughed buccal epithelial cells can be used to genotype persons. This has found application in forensics (Roy et al. 1997) and may be ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL 403 Linkages with General Health used for diagnostic purposes in the future (van Schie and Wilson 1997). Saliva has the potential of replacing blood, the current standard for testing many diseases and conditions (e.g., diabetes, infectious disease, Parkinson's disease, alcoholic cirrhosis, Sjogren's syndrome, and cystic fibrosis sarcoidosis). Important goals for the future are the development of new diagnostic tests for early disease detection, defining individual patient risk of adverse response to drugs, monitoring therapeutic progress, and determin- ing outcomes of treatment. Key issues in the development of a new generation of saliva diag- nostics include their selectivity, sensitivity, response time, dynamic range (values of interest), rep- resentative sampling, and, perhaps most important, their reliability or stability as well their ability to assess multiple substances simultaneously. Conclusion For the clinician the mouth and face provide ready access to physical signs and symptoms of local and generalized disease and risk factor exposure. These signs and symptoms augment other clinical features of underlying conditions. Comprehensive care of the patient requires knowledge of these signs and symp- toms, their role in the clinical spectrum of general diseases and conditions, and their appropriate man- agement. Oral biomarkers and surrogate measures are also being explored as means of early diagnosis. With further development and refinement, oral- based diagnostics such as salivary tests can be- come widely used and acceptable tools for individu- als, health care professionals, researchers, and com- munity programs. The continued refinement of imaging techniques also has the potential of using oral imaging to identify early signs of skeletal bone degeneration. THE MOUTH AS A PORTAL OF ENTRY FOR INFECTION Chapter 3 provides an overview of the effects of oral microbial infections with viruses, bacteria, and fungi. More than 500 bacterial strains have been identified in dental biofilm, and more than 150 bacterial strains have been isolated from dental pulp infections. More recently, 37 unique and previously unknown strains of bacteria were identified in dental plaque (biofilm) (Kroes et al. 1999). Most oral lesions are opportunis- tic infections, that is, they are caused by microorgan- isms commonly found in the mouth, but normally kept in check by the body’s defense mechanisms. These microorganisms can induce extensive local- ized infections that compromise general well-being in and of themselves. However, they also may spread to other parts of the body if normal barriers are breached. The oral mucosa is one such barrier that provides critical defense against pathogens and other challenges (Schubert et al. 1999). Salivary secretions are a second major line of defense. Damage to the oral mucosa from mechanical trauma, infection, or salivary dysfunction with resulting derangements in lubricatory and antimicrobial functions of saliva, as a result of chemotherapy, radiation, and medications causing hyposalivation, allows a portal of entry for invading pathogens. Oral Infections and Bacteremia Oral microorganisms and cytotoxic by-products associated with local infections can enter the blood- stream or lymphatic system and cause damage or potentiate an inappropriate immune response else- where in the body. Dissemination of oral bacteria into the bloodstream (bacteremia) can occur after most invasive dental procedures, including tooth extrac- tions, endodontic therapy, periodontal surgery, and scaling and root planing. Even routine oral hygiene procedures such as daily toothbrushing, subgingival irrigation, and flossing may cause bacteremia. However, these distant infections have been seen more often in high-risk patients such as those who are immunocompromised. Oral bacteria have several mechanisms by which they invade mucosal tissues, perhaps contributing to their ability to cause bacteremias. For example, oral bacteria and their products may invade the periodontal tissues directly. Actinobacillus actino- mycetemcomitans has been found in gingival connective tissue in patients with localized juvenile periodontitis (Christersson et al. 1987a,b, Meyer et al. 1991, Riviere et al. 1991). Invasion of tissue by Porphyromonas gingivalis has also been described in vivo (Saglie et al. 1988) and in vitro (Njoroge et al. 1997, Sandros et al. 1993, 1994, Weinberg et al. 1997). Although oral bacteria can enter the blood through injured or ulcerated tissue, bacterial invasion of periodontal tissues represents another possible mechanism. In the immunocompetent individual, bacteremia originating from the oral cavity is usually transient and harmless. However, if the individual’s immune system is compromised, the normally harmless oral bacteria may pose a significant risk. The morbidity and mortality associated with oral foci of infections are hard to assess. This is due to the formidable task 104 ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL of tracking the source of an infection unless the responsible pathogen is indigenous to a specific anatomic location. Viridans group streptococci (VGS) have a low degree of virulence but can be associated with mor- bidity and mortality under certain circumstances. Increased pathogenicity of Streptococcus viridans is most prominent in individuals with neutropenia (low blood counts of circulating white blood cells called neutrophils) and has been associated with a toxic- shock-like syndrome (TSLS) or viridans streptococ- cal shock syndrome (VSSS), as well as with adult res- piratory distress syndrome (ARDS) (Bochud et al. 1994). Although a high degree of morbidity is associat- ed with viridans streptococcal bacteremia, a low inci- dence of mortality has been reported (Heimdahl et al. 1989). Several studies have shown that under adverse circumstances oral flora and oral infections are associated with increased incidence of morbidity and even mortality (Engelhard et al. 1995, Lucas et al. 1998, Martino et al. 1995, Ruescher et al. 1998, Sparrelid et al. 1998, Sriskandan et al. 1995). Reduction of oral foci of infection decreases systemic complications, specifically in severely neutropenic patients undergoing chemotherapy (Heimdahl et al. 1984). In addition, hospital stays for patients with oral mucositis undergoing autologous bone marrow transplants were longer than for those without oral mucositis (Ruescher et al. 1998). Other cohorts identified at increased risk far sys- temic complications due to oral bacteria include hos- pitalized patients unable to perform adequate oral hygiene, those receiving saliva-reducing medications, and those taking antibiotics that alter the oral flora. A positive dental plaque culture for aerobic pathogens was significantly associated with the development of hospital-acquired pneumonia and bacteremia in a study of individuals in an intensive care unit (ICU) (Fourrier et al. 1998). In addition, several case reports have been pub- lished implicating indigenous oral flora in the devel- opment of brain abscesses (Andersen and Horton 1990, Andrews and Farnham 1990, Baker et al. 1999, Gallagher et al. 1981, Goteiner et al. 1982, Saal et al. 1988). This serious condition is associated with a mortality rate of almost 20 percent and full recovery in only slightly more than 50 percent of all patients (Goteiner et al. 1982). These data are based on single case reports and most probably represent rare events. However, they provide additional examples that point to the potential pathogenicity of the normal oral flora during special adverse circumstances. Linkages with General Health Oral Infections as a Result of Therapy Chemotherapy Oral mucositis can be a major dose-limiting problem during chemotherapy with some anticancer drugs, such as 5-fluorouracil, methotrexate, and doxoru- bicin. It is estimated that approximately 400,000 patients undergoing cancer therapy each year will develop oral complications (NIH 1990). Infection of ulcerated mucous membranes often occurs after chemotherapy, especially since patients are usually immunocompromised. Bacterial, fungal, and viral causes of mucositis have been identified (Feld 1997). The mechanism by which cancer-chemotherapy- induced mucositis occurs is likely associated with the rapid rate of turnover of oral epithelial cells. In addi- tion, other components likely include upregulation of pro-inflammatory cytokines and metabolic by- products of colonizing oral microflora (Sonis 1998). Chemotherapy alters the integrity of the mucosa and contributes to acute and chronic changes in oral tis- sue and physiologic processes (Carl 1995). The ulcerated mucosa is susceptible to infection by microbial flora that normally inhabit the oral cavity, as well as by exogenous organisms, and exacerbates the existing mucositis. Further, these microflora can disseminate systemically (Pizzo et al. 1993, Rolston and Bodey 1993). Compromised salivary function can further elevate risk for systemic infection of oral origin. Both indigenous oral flora and hospital-acquired pathogens have been associated with bacteremias and systemic infection (Schubert et al. 1999). Changes in infection profiles in myelosuppressed (immunosup- pressed) cancer patients tend to occur in cyclic fash- ion over many years. This evolving epidemiology is caused by multiple factors including use of antibi- otics. Gram-positive organisms including viridans streptococci and enterococci are currently associated with systemic infection of oral origin in myelosup- pressed cancer patients. In addition, gram-negative pathogens including P. aeruginosa, Neisseria spp., and Escherichia coli remain of concern. Cancer patients undergoing bone marrow radia- tion who have chronic periodontal disease may also develop acute periodontal infections with systemic complications (Peterson et al. 1987). The extensive ulceration of gingival sulcular epithelium associated with periodontal disease is not directly observable clinically, yet may represent a source for disseminat- ed infection by an extensive array of organisms. Inflammatory signs may be masked due to the under- lying bone marrow suppression. ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL 105 Linkages with General Health Viruses are also associated with clinically impor- tant oral disease in patients receiving chemotherapy (Rolston and Bodey 1993, Pizzo et al. 1993). Infections caused by herpes simplex virus, varicella- zoster virus, and Epstein-Barr virus typically result from reactivation of a latent virus, whereas cytomegalovirus infections can result via reactivation of a latent virus or a newly acquired virus. The sever- ity of the infection, including fatal outcome, depends on the degree of immunocompromise. Many agents and protocols have been investigat- ed to manage or prevent mucositis (Peterson 1999, Schubert et al. 1998). For example, various biologic response modifiers, including transforming growth factor B3 or keratinocyte growth factor, have been under recent study in randomized clinical trials. Allopurinol mouthwash and vitamin E have been cited as agents that can decrease the severity of mucositis, although more extensive testing is neces- sary. Prostaglandin E2 was not shown to be effective in prophylaxis of oral mucositis following bone mar- row transplant; however, more recent studies indi- cate possible efficacy when administered via a differ- ent dosing protocol. Oral cryotherapy appears to be efficacious in reducing severity of oral mucositis caused by 5-fluorouracil and related compounds (Rocke et al. 1993). Local application of capsaicin preparations may be effective in controlling oral mucositis pain as dis- tinguished from tissue injury itself (Berger et al. 1995). Capsaicin and its analogs are the active ingre- dients in chili peppers. Capsaicin’s clinical potential derives from the fact that it elevates the threshold for pain in areas to which it is applied. Radiation Therapy Radiation therapy disrupts cell division in healthy tissue as well as in tumors and also affects the normal structure and function of craniofacial tissues, includ- ing the oral mucosa, salivary glands, and bone. Oral- facial complications are common after radiation ther- apy to the head and neck. The most frequent, and often the most distressing, complication is mucositis, but adverse reactions can affect all oral-facial tissues (Scully and Epstein 1996). Radiation can cause irreversible damage to the salivary glands, resulting in dramatic increases in dental caries. Oral mucosal alterations may become portals for invasion by pathogens, which may be life- threatening to immunosuppressed or bone-marrow- suppressed patients. A less common but very serious adverse consequence is destruction of bone cells and bone death, called osteoradionecrosis (ORN). ORN can result in infection of the bone and soft tissue and can require surgery to excise the dead tissue, which can in turn leave the face badly disfigured as well as functionally impaired (Field et al. 2000). The likeli- hood of ORN is increased with trauma to the bone, including that caused by tooth extraction (Murray et al. 1980a, b). The risk is especially marked when the trauma occurs near the time of radiation (Epstein et al. 1987). Management includes elimination of acute or potential dental and periodontal foci of disease, increased patient participation in oral hygiene, use of oral topical fluorides for caries prevention, and use of antiviral, antifungal, or antimicrobial therapy for management of infections associated with mucositis. Combined Cancer Therapies Rapid developments have occurred in the use of blood cell growth factors for treatment of various conditions, including the anemia of end-stage renal disease, the neutropenia occurring with cancer care, and the bone marrow toxicity and mucositis that can follow aggressive chemotherapy or radiation therapy (Sonis et al. 1997, Williams and Quesenberry 1992). Sonis et al. (1997) found that topical application of transforming growth factor beta (TGF-B) in the ham- ster model of oral mucositis significantly reduced basal cell proliferation and reduced the severity of mucositis associated with 5-fluorouracil treatment. Other growth factors considered for use in reducing mucositis include granulocyte-monocyte colony-stimulating factor and granulocyte colony- stimulating factor. Bone morphogenetic proteins are also in development for alleviating the toxicity and mucositis that follow chemotherapy and radiation therapy. Other approaches to reducing mucositis and adverse oral effects of chemotherapy and radiation therapy include fractionating the dose of radiation, and combining chemotherapy with growth factors or with less toxic oncostatic agents. Although the oral mucositis occurring in chemotherapy and in head and neck radiation patients shares many characteristics, distinct differences also exist (NIH 1990, Schubert et al. 1998, Wilkes 1998). For example, in contrast to chemotherapy-associated lesions, radiation damage is anatomically site- specific, toxicity is localized to irradiated tissue volumes. The degree of damage depends on treatment-regimen-related factors, including the type of radiation used, the total dose administered, the fractionation, and field size. Thus, research involving both cohorts of cancer patients remains essential to enhancing patient management. Development of new technologies to prevent cancer-therapy-induced oral mucositis could 106 ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL substantially reduce the risk for oral and systemic infections, oral pain, and the number of hospital days. Improvement in quality of life and reduction in health costs are also likely and desirable outcomes. The new technologies could also provide a set- ting in which novel classes of chemotherapeutic drugs, utilized at increased doses, could be imple- mented. These advances in turn could lead to enhanced cancer patient survival and lengthen the duration of disease remission. Pharmaceuticals A number of medications used to treat systemic dis- eases can cause oral complications, ranging from xerostomic effects to alterations in the surface struc- ture of the enamel or mucosa. More than 400 over- the-counter and prescription drugs have xerostomic side effects (Sreebny and Schwartz 1997). These include tricyclic antidepressants, antihistamines, and diuretics. The dimensions and impact of these side effects vary depending on the response of the indi- vidual patient and the duration of medication use. Staining of the teeth or mucosa is associated with a variety of drugs, including tranquilizers, oral con- traceptives, and antimalarials. The antibiotic tetracy- cline can cause enamel hypoplasia when taken by the mother during pregnancy and by children during tooth development. The antimicrobial mouthrinse agent chlorhexidine also can stain the teeth, but this staining is external and can be removed by dental prophylaxis. Other drugs have been associated with gingival overgrowth, including cyclosporin, which has been used as an immunosuppressant in the United States since 1984 to prevent rejection of transplanted organs and bone marrow. This drug has also been used in other countries for treatment of type 2 dia- betes, rheumatoid arthritis, psoriasis, multiple sclero- sis, malaria, sarcoidosis, and several other diseases with an immunological basis (Adams and Davies 1984). Other drugs that cause gingival overgrowth include calcium ion channel blocking agents used in the treatment of angina pectoris and postmyocardial syndrome, such as nifedipine and verapamil (Lucas et al. 1985), and phenytoin (sodium 5,5-phenylhy- dantoin), used in the treatment of epilepsy and also for management of other neurological disorders. Treatment often consists of using an alternate drug, although this is not always possible. Conservative periodontal therapy can reduce the inflammatory component of enlargement, however, surgery is often required. Oral candidiasis is typically caused by opportunistic overgrowth of Candida albicans. Drugs that cause systemic bone marrow suppression, oral Linkages with General Health mucosal injury, or salivary compromise collectively promote the risk for clinical infection. In addition, antibiotics and concurrent steroid therapy often alter oral flora, thereby creating an environment for fungal overgrowth. In high-risk cancer patients, fungal infection can cause severe morbidity and even death. Infective Endocarditis The purported connection between oral infection and a specific heart disease, infective endocarditis, has a long history. Endocarditis is caused by bacteria that adhere to damaged or otherwise receptive sur- faces of the tissue that lines heart valves (the endo- cardium) (Weinstein and Schlesinger 1974). Dental and other surgical procedures may predispose sus- ceptible patients to infective endocarditis by induc- ing bacteremias (Lacassin et al. 1995). However, bac- teremias from oral infections that occur frequently during normal daily activites, coincidental even with chewing food, toothbrushing, and flossing, con- tribute more substantially to the risk of infective endocarditis (Bayliss et al. 1983, Dajani et al. 1997, Strom et al. 1998). Oral organisms are common etio- logic agents of infective endocarditis (Bayliss et al. 1983). For example, strains of S. sanguis, as well as gram-negative oral bacteria including Haemophilus aphrophilus, A. actinomycetemcomitans, E. corrodens, Capnocytophaga spp.. and Fusobacterium nucleatum, have been associated with bacterial endocarditis (Barco 1991, Geraci and Wilson 1982, Kaye 1994, Moulsdale et al. 1980). Infective endocarditis occurs with different incu- bation periods, which differ in causative bacteria and signs and symptoms. For example, Staphylococcus aureus endocarditis may have a rapid onset and fatal course if it affects the left side of the heart. With a more indolent course, patients may often be unaware of infection and may experience fever, night chills, myalgia, and arthralgia for a considerable period of time before diagnosis. The infection is often curable if diagnosed and treated early. The classic risk factors for endocarditis include cardiac valve disorders (valvulopathies) that include theumatic and congenital heart disease, complex cyanotic heart disease in children, and mitral valve prolapse with regurgitation. Recent studies indicate that the use of certain diet drugs (fenfluramine and dexfenfluramine) has induced cardiac valvulopathy, which may in some cases be transient. Among at-risk persons, bacteremias are more likely to occur in those with periodontal disease (Silver et al. 1977). However, the oral pathogens causing periodontitis have only rarely been shown to cause endocarditis. ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL 107 Linkages with General Health Prevention of infective endocarditis from oral bacteria depends on limiting the entry and dissemi- nation of bacteria through the bloodstream and lym- phatic circulation. Antibiotic prophylaxis for dental procedures that are likely to provoke bacteremia has historically been recommended (Dajani et al. 1997, Durack 1995). A recent study, however, suggests that receiving dental treatment does not significantly increase the risk of infective endocarditis, even in patients with valvular abnormalities (Strom at al. 1998). Further research is necessary to determine whether some heart or valvular conditions or certain dental procedures, such as surgery or scaling, would require coverage with pre-procedural antibiotics and others would be precluded. Oral Infections and Respiratory Disease Pathogens in the oral cavity can also gain access to the airway, sometimes with serious consequences. In adults, bacterial pneumonias are strongly associated with aspiration of bacteria into the lower respiratory tract, which is normally sterile. Common respiratory pathogens such as Streptococcus pneumoniae, Streptococcus pyogenes, Mycoplasma pneumoniae, and Haemophilus influenzae can colonize the oropharynx and the lower airway. In addition, oral bacteria including A. actinomycetemcomitans (Yuan et al. 1992), Actinomyces israelii (Morris and Sewell 1994, Zijlstra et al. 1992), Capnocytophaga spp. (Lorenz and Weiss 1994), Eikenella corrodens (Joshi et al. 1991), Prevotella intermedia, and Streptococcus con- stellatus (Shinzato and Saito 1994) can be aspirated into the lower airways (Scannapieco 1998, 1999). Chronic obstructive pulmonary disease, charac- terized by obstruction of airflow due to chronic bron- chitis or emphysema and by recurrent episodes of respiratory infection, has been associated with poor oral health status (Hayes et al. 1998, Scannapieco et al. 1998). A positive relationship between periodon- tal disease and bacterial pneumonia has been shown by Scannapieco and Mylotte (1996). Although oral bacteria, including periodontal pathogens, have the potential for causing respiratory infections, the frequency and nature of such infec- tions are not known and merit further study. Oral Transmission of Infections Besides being a portal of entry for infections, the mouth is an important source of potentially patho- genic organisms and is often the vehicle by which infection is delivered to the bodies of others. Microorganisms were not discovered in the mouth until the seventeenth century, when van Leewenhoek examined dental plaque using a microscope he had constructed. In 1884, Koch demonstrated that tuberculosis could be transmitted by airborne droplets from the mouth and respiratory tract. Since that time, we have learned that many common respi- ratory infections, such as influenza, the common cold, pneumonia, and tuberculosis, can be transmit- ted from oral secretions. Before the development of effective vaccines, orally transmitted diseases such as chickenpox, measles, mumps, polio, and diphtheria were a major source of morbidity and mortality in childhood. Viral diseases such as hepatitis B, herpes labialis, acute herpetic gingivostomatitis, cyto- megalovirus, and infectious mononucleosis may also originate from oral contact. Disease-causing microorganisms can be spread by direct contact (with saliva or blood from the mouth) or indirect contact (with saliva- or blood- contaminated surfaces, including hands or lips), droplet infection (from coughing, sneezing, or even normal speech), or by aerosolized organisms. These organisms can be inhaled, ingested, or taken in through mucous membranes in the eyes, nose, or mouth or through breaks in the skin. A number of diseases can be spread via oral sexual contact, includ- ing gonorrhea, syphilis, trichomoniasis, chlamydia, and mononucleosis. As mentioned earlier, the oral mucosa and saliva provide significant defense against disease transmis- sion. Epidemiological and animal studies are provid- ing evidence, however, that the oral cavity may be the site for transmission of serious systemic infections despite the protective factors in saliva (see Chapter 2). Infection with HIV provides a case in point (Baba et al. 1996, Dillon et al. 2000, Pope et al. 1997, Ruprecht et al. 1999, Stahl-Hennig et al. 1999, Baron et al. 2000). Early in the 1980s, when AIDS was first identi- fied in the United States, concern was expressed about casual (i.e., nonsexual) transmission of HIV (CDC 1983, 1985). Detailed household studies did not demonstrate transmission of HIV, even when family members shared eating utensils and tooth- brushes with an HIV-affected member (Fischl et al. 1987, Rogers et al. 1990, Sande 1986). Similarly, sur- veillance data collected over time showed no evi- dence of casual transmission (Ward and Duchin 1997). Only one nonoccupational episode of HIV trans- mission has been attributed to blood-contaminated saliva (CDC 1997): this incident involved intimate kissing between sexual partners. There have been a few cases of HIV transmission from performing oral 108 ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL sex on a person infected with HIV, and it is also pos- sible to become infected with HIV by receiving oral sex. In the San Francisco Options Study of men who have sex with men identified within 12 months of HIV seroconversion, oral transmission represented 7.8 percent of primary HIV infections (Dillon et al. 2000). Rothenberg et al. (1998) reviewed epidemio- logic studies and reports of 38 cases of oral transmis- sion of HIV in the literature. They concluded that although oral-genital contact may be less efficient than needle-sharing or anal intercourse for the trans- mission of HIV, its increased use by men who have sex with men (Ostrow and DiFranceisco 1996, Schwarcz et al. 1995) and in crack cocaine smokers (Faruque et al. 1996a,b) may increase its contribu- tion to HIV transmission over time. Several studies provide evidence that when the oral environment is compromised, the mouth can be a potential site of transmission of infectious microbes. Data from Faruque et al. (1996a,b) and Wallace et al. (1996) suggest that there is a positive association between the presence of oral lesions resulting from crack co- caine use, receptive oral intercourse, and HIV trans- mission. A case report has documented the passage of HIV from a partner who is HIV-positive to one who is HIV-negative in the presence of periodontal disease but in the absence of other risk factors (Padian and Glass 1997). Because the type, duration, and. fre- quency of oral contact in past studies may not have been specified, the risk could be somewhat higher for oral transmission of HIV than previously reported. The risk might also vary depending on factors such as viral load, infectious dose, area of exposure, and presence or absence of oral lesions. Additional stud- ies are needed to evaluate the risk of oral-genital transmission of HIV; some are under way JQ. Greenspan, K. Page-Schafer, personal communica- tion, 1999). Other sexually transmitted diseases (STDs) can occur through oral contact. For example, pharyngeal infection with Chlamydia trachomatis has been found in 3 to 6 percent of men and women attending STD clinics. Most infections are asymptomatic (Holmes et al. 1999). Another common sexually transmitted infection, herpes simplex virus, commonly infects the pharynx and is seen in 20 percent of patients with primary genital herpes. The painless chancre of pri- mary syphilis can be found in the oral cavity; howev- er, there are no data on the prevalence of this site of infection for Treponema pallidum. Among persons with gonorrhea, pharyngeal infection occurs in 3to7 percent of heterosexual men, 10 to 20 percent of het- erosexual women, and 10 to 25 percent of men who have sex with men (Holmes et al. 1999). Gonococcal Linkages with General Health infection can cause acute pharyngitis, but is usually asymptomatic. The transmission of pharyngeal gon- orthea to sex partners had been thought to be rare. However, in one study, 17 of 66 men who had sex with men who had urethral gonorrhea reported insertive oral sex as their only risk factor in the past 2 months (Lafferty et al. 1997). Conclusion The role of the mouth as a portal of entry for infec- tion presents ever-new challenges for study. Although oral tissues and fluids normally provide significant barriers and protection against microbial infections, at times these infections can not only cause local dis- ease but, under certain circumstances, can dissemi- nate to cause infections in other parts of the body. The control of existing oral infections is clearly of intrinsic importance and a necessary precaution to prevent systemic complications. ASSOCIATIONS AMONG ORAL INFECTIONS AND DIABETES, HEART DISEASE/STROKE, AND ADVERSE PREGNANCY OUTCOMES Recent studies have reported associations between oral infections—primarily periodontal infections— and diabetes, heart disease and stroke, and adverse pregnancy outcomes, but sufficient evidence does not yet exist to conclude that one leads to the other. This section characterizes the nature of these associ- ations by describing the quality of the evidence sup- porting the reports. Both observational and experi- mental studies were accepted as admissible evidence. Table 5.3 presents the hierarchy of evidence used to interpret these associations. Where there are opera- tive mechanisms proposed that support an associa- tion between oral infectious agents and the systemic conditions in question, they are introduced at the outset. These are followed by animal studies and then by epidemiologic or population-based studies. The evidence for each association is presented in the table in rank order according to the rigor of the study design. The Periodontal Disease—Diabetes Connection There is growing acceptance that diabetes is associat- ed with increased occurrence and progression of peri- odontitis—so much so that periodontitis has been called the “sixth complication of diabetes” (Lde 1993). ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL 109 Linkages with General Health The risk is independent of whether the diabetes is type l or type 2. Type 1 diabetes is the condition in which the pancreas produces little or no insulin. It usually begins in childhood or adoles- cence. In type 2 diabetes, secretion and utilization of insulin are impaired; onset is typically after age 30. Together, these two types of diabetes affect an esti- mated 15.7 million people in the United States and represent the seventh leading cause of death (NIDDK 1999). The goal of diabetic care is to lower blood glu- cose levels to recommended levels. Some investiga- tors have reported a two-way connection between diabetes and periodontal disease, proposing that not only are diabetic patients more susceptible to peri- odontal disease, but the presence of periodontal dis- ease affects glycemic control. This section explores the bidirectional relationship, beginning with the effects of diabetes on periodontal disease. Effects of Diabetes on Periodontitis Prevalence and Severity Several reviews have described candidate mecha- nisms to explain why individuals with diabetes may be more susceptible to periodontitis (Grossi and Genco 1998, Manouchehr-Pour and Bissada 1983, Murrah 1985, Oliver and Tervonen 1994, Salvi et al. 1997, Wilton et al. 1988). These include vascular changes, alterations in gingival crevicular fluid, alter- ations in connective tissue metabolism, altered host immunological and inflammatory response, altered subgingival microflora, and hereditary: patterns. Studies were classified by type of diabetes and age of study population (see Table 5.4). Type 1 Diabetes. Ten reports focused principally on children and adolescents with type 1 diabetes, com- paring them with groups of similar ages without dia- betes (Cianciola et al. 1982, de Pommereau et al. 1992, Faulconbridge et al. 1981, Firatli 1997, Firatli et al. 1996, Goteiner et al. 1986, Harrison and Bowen 1987, Novaes et al. 1991, Pinson et al. 1995, Ringelberg et al. 1977). All but one of the studies (Goteiner et al. 1986) reported greater prevalence, extent, or severity of at least one measure or index of periodontal disease (e.g., gingival inflammation, probing pocket depth, loss of periodontal attach- ment, or radiographic evidence of alveolar bone loss) among subjects with diabetes, even though these investigations were conducted ina variety of coun- tries across several continents. Another set of studies on the relationship between type 1 diabetes and periodontal disease included subjects with and without diabetes between the ages of 15 and 35 (Cohen et al. 1970, Galea et al. 1986, Guven et al. 1996, Kjellman et al. 1970, Rylander et al. 1987, Sznajder et al. 1978). All six studies reported greater prevalence, extent, or sever- ity of at least one measure or index of periodontal disease. A third set of studies conducted in Scandinavia looked at the relationship between periodontal dis- ease and type 1 diabetes (or diabetes reported as requiring insulin therapy without specification of diabetes type) in adults between 20 and 70 years old. Three of the four studies were cross-sectional (Glavind et al. 1968, Hugoson et al. 1989, Thorstensson and Hugoson 1993), and one was a treatment follow-up study (Tervonen and TABLE 5.3 Hierarchy of evidence used in analyzing and interpreting results Quality of Evidence Strength of Recommendation A: There is good evidence to support the recommendation. There is fair evidence to support the recommendation. moO |; Evidence obtained from at least one properly randomized controlled trial. Il-1: Evidence obtained from well-designed controlled trials without randomization. Il-2: Evidence obtained from well-designed cohort of case-control analytic studies, preferably from more than one center or research group. Il-3: Evidence obtained from multiple time series with or without the intervention. Dramatic results in uncontrolled experiments (such as the results of the introduction of penicillin treatment in the 1940s) could also be regarded as this type of evidence. Ill: Opinions of respected authorities, based on clinical experience; descriptive studies and case reports; or reports of expert committees. There is insufficient evidence to recommend for or against, but recommendations may be made on other grounds. There is fair evidence to support the recommendation that the intervention be excluded. There is good evidence to support the recommendation that the intervention be excluded. Source: Adapted from U.S. Preventive Services Task Force 1996, ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL Linkages with General Health TABLE 5.4 Summary of studies of the association between diabetes and periodontal diseases, classified by strength of evidence, diabetes type, and age group Number of Measure of Subjects Ages? Periodontal Other Study Diabetes a. Diabetes a.Diabetes Disease Status: Diabetes-Related Evidence Country Design Type b. Control b. Control Diabetes Effect: Variables Considered Level! Firatli 1997 Turkey Prospective 1 a.44 a.12.2(mean) — Ging: 0s Glycemic control U-2 b.20 b.12.3 (mean) — Ppd:0s Duration of diabetes Lpa: 1s Cohen et al. 1970 USA Prospective a a.21 a. 18,35 Ging: 1s None Il-2 b. 18 b.18, 35 Lpa: 11, 1s Tervonen and Finland Prospective 1 a.36 a. 24, 36 Ging: 0e Glycemic contral I-2 Karjalainen 1997 b.10 b. 24, 36 Pod: Ir Duration of diabetes Lpa: le Diabetes complications Novaes et al. 1996 Brazil Prospective 2 a.30 a. 30,77 Ppd: 1s, ir Glycemic control I-2 b.30 b.30, 67 Lpa, 15, Ir Nelson et al. 1990 USA Prospective 2 a.720 a.15,55+ XRBL: 1i, 1p None Il-2 b. 1,553 b. 15, 55+ Taylor et al. 1998a USA Prospective 2 a.24 a. 15,57 XRBL: i, Ir None Il-2 b.338 b.15,57 Taylor et al. 19986 USA Prospective 2 a2l a. 15,49 XRBL: ti, IF Glycemic control i-2 b.338 b. 15,49 Goteiner et al. 1986 USA Cross-sectional = 1 a. 169 a.schoal ages Ging: 0s None il b.80 b.5, 18 Lpa: Op, Os PDI: 0s Harrison and USA Cross-sectional 1 a.30 a.4,19 Ging: 1s Glycemic control il Bowen 1987 b.30 b.4,19 Lpa: 1p Novaes et al. 1991 Brazil Cross-sectional 1 a.30 a.5,18 Ging: 1s None Ml “b.30 b.5,18 Ppd: Os XRBL: 1s Cianciola et al. 1982 USA Cross-sectional 1 a. 263 a.<10,>19 Ging: Ip Duration of diabetes IN b.208 b.<10,>19 Lpa: Ip XRBL: 1p, 1s JPS: 1p,1s de Pommereau France —- Cross-sectional. «= a.85 a. 12, 18 Ging: le Glycemic control tll et al. 1992 b.38 b. 12, 18 Lpa: Oe, Op, Os Duration of diabetes XRBL: Oe, Op, 0s Ringelberg USA Cross-sectional 1 a.56 a. 10, 16 Ging: 1s None Ml et al. 1977 b.41 b.10, 12 MGI: 1s Firatli et al. 1996 Turkey Cross-sectional 1 a.77 a.12.5(mean) — Ging: 0s Duration of diabetes Ml b.77 b.12.6(mean} — Ppd: 1s Lpa: is Pinson et al. 1995 USA Cross-sectional 1 a.26 a.7-18 Ging: 1s Glycemic control Ml b.24 b.7-18 Ppd: 0s Duration of diabetes Lpa:Qs Faulconbridge England Cross-sectional a.94 a.5,17 Ging: 1s Duration of diabetes Mil et al. 1981 b.94 b.5,17 Kjeliman et al. 1970 Sweden Cross-sectional = 1* a.105 a. 15,24 Ging: le Glycemic control lit b.52 b.15, 24 Ppd: Os Diabetes complications XRBL: 0s Guven et al. 1996 Turkey Cross-sectional 1 a.10 a. 18,27 Ging: le None til (continues) ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL 111 Linkages with General Health TABLE 5.4 continued Number of Measure of Subjects Ages? Periodontal Other Study Diabetes a.Diabetes a. Diabetes Disease Status: Diabetes-Related Evidence Country Design Types b. Control b. Control Diabetes Effect: Variables Considered Level Rylander et al. 1987 Sweden Cross-sectional = a.46 a. 18, 26 Ging: le, 1p Diabetes complications tit b.41 b, 19,25 Ppd:0e Lpa: le, ip XRBL:Qp Sznajder et al. 1978 Argentina Cross-sectional = -1* a.20 a.9, 29 Ging: 1s None il b.26 b.9,29 Lpa: 0s Galea et al. 1986 Malta Cross-sectional = -1* a.82 a.5,29 Ppd: 1p Glycemic control tll b.unknown — b.5,29 Duration of diabetes Diabetes complications Hugoson et al. 1989 Sweden Cross-sectional = a. 154 a. 20,70 Ging: le Duration of diabetes Ill b.77 b. 20,70 Ppd: 1e, Ip, 1s XRBL: 1s Glavind et al. 1968 Denmark Cross-sectional 1* a.51 a.20,40 Ging: 0s Duration of diabetes II" b.51 b. 20,40 Ppd: Os Diabetes complications Lpa: 1s XRBL: 1s Thorstensson and Sweden Cross-sectional 1 a.117 a.40, 70 Ging: 0e Duration of diabetes ill Hugoson 1993 b.99 b.40,70 Ppd: le, 1s Onset age XRBL: 1s Morton et al. 1995 Mauritius Cross-sectional 2 a.24 a. 26,76 Ging: 1p None lil b.24 b.25, 73 Ppd: 1s Lpa: 1s Shlassman et al.1990 USA Cross-sectional 2 a, 736 a.5,45+ Lpa: Ip None ill b, 2,483 b.5,45+ XRBL: Ip Emrich et al. 1991 USA Cross-sectional 2 a.254 a.15,55+ Lpa: tp, 1s None ill b. 1,088 b.15,55+ XRBL: 1p, 1s Wolf 1977 Finland Cross-sectional = 1,2 a. 186 a. 16,60 Ging: 1s Glycemic control Ml b. 156 b. 16,60 Lpa: 1s Duration of diabetes XRBL: 1s Diabetes complications Benveniste USA Cross-sectional §=—-1,2* a.53 a.5,72 Ging: 0s None Itl et al. 1967 b.71 b.5,72° Ppd: Op, Os Finestone and USA Cross-sectional 1,2* a. 189 a. 20,79 Pl: 4s Glycemic control Mtl Boorujy 1967 b.64 b.20, 79 Duration of diabetes Diabetes complications Belting et al. 1964 USA Cross-sectional 1,2" a.78 a.20,79 Pl: 1s Diabetes severity ill b.79 b.20,79 Oliver and Tervonen USA Cross-sectional 1,2 a.114 a.20, 64 Ppd: Te, 1p None i 1993 b, 15,132 b.20, 64 Lpa: 1e, Op, Os Yavuzyilmaz et al. Turkey Cross-sectional ~=—-1,2 a1? a.25,74 Ppd: 1s None tll 1996 b.17 b, 19,29 Bridges et al. 1996 USA Cross-sectional 1,2 a.118 a. 24,78 Ging: 0s Glycemic control ill b.115 b. 24, 78 Ppd: Os Duration of diabetes Lpa: 1s Sandler and Stahl USA Cross-sectional —-1,2* a. 100 a. 20,69 PDR: le None lil 1960 b. 3,894 b. 20,69 Bacic et al. 1988 Yugoslavia Cross-sectional 1,2 a.222 a.<20, 60+ Pod: 1e, 1p, 1s Glycemic control lll b. 189 b. <20,60+ Duration of diabetes Diabetes complications Hove and Stallard USA Cross-sectional —1,2* a. 28 a.20,40+ Ging: 0s Duration of diabetes ill 1970 b. 16 b. 20, 40+ Ppd: Os Diabetes severity XRBL:0s 112 ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL Linkages with General Health Number of Measure of Subjects Ages? Periodontal Other Study Diabetes a.Diabetes a. Diabetes Disease Status: Diabetes-Related Evidence Country Design Type? b. Control b.Control Diabetes Effect: Variables Considered Leveld Mackenzie and USA Cross-sectional 9 a. 124 a. 32,78 XRBL: 0s None It Millard 1963 b.92 b.32, 78 Sznajder et al. 1978 Argentina Cross-sectional 9 a.63 a. 30,49 Ging: 1s None Mt b.39 b.30, 50 Lpa: 1s Dolan et al. 1997 USA Cross-sectional = 9 Weighted a.45,75+ Lpa: 1e, 1p, 1s None itt a.107 b.45, 75+ b.554 Grossi et al. 1994 USA Cross-sectional 9 a. 1,426 All: 25, 74; Lpa: 1s, ip None Ml b.69 unknown for diabetes Tervonen and Finland Cross-sectional © 9 a.50 a. <30,40+ Ging: le Glycemic control Ml Knuuttila 1986 . b.53 b.<30,40+ Pod: 1e, 1p XRBL: 0s Campbell 1972 Australia Cross-sectional = 9 a.70 a. 17,39 Pl: Ip, 15 None il b. 102 b. 17,39 Aibrecht et al. 1988 Hungary Cross-sectional 9 a. 1,360 a.15,65+ Ging: 1s None it b.625 b.15,65+ PI: 0s Szpunar et al. USA Cross-sectional 9 a.474 a.6,65+ Pl: 1s None I 1989 (NHANES {) b. 15,174 b.6,65+ Szpunar et al. USA Cross-sectional 9 a.322 a. 15,65+ Pl: 1s, None Ul 1989 (HHANES) b.8,040 b.12,65+ dLevels of evidence are delineated in Table 5.3. aDiabetes type: 1 = type 1 diabetes mellitus; 2 = type 2 diabetes mellitus; 1,2 = both type 1 and type 2 diabetes mellitus; 9 = diabetes type not specified and not clearly ascertaina- ble from other information in the report; * = diabetes type not specified but ascertained by reviewer from other information in the report. bAges: subjects’ ages presented as minimum, maximum reported for those with diabetes and controls unless otherwise specified. < Measure of periodontal disease status. Measures used include Ging = gifigivitis or gingival bleeding; Ppd = probing pocket depth; Lpa = loss of periodontal attachment; XRBL = radiographic bone loss; JPS = juvenile periodontal score; MGl = madified gingival index; Pi = Russell's periodontal index; POR = periodontal disease rate (proportion of teeth affected by periodontal disease), The number following the measure corresponds to greater disease in those with diabetes (1) or no difference between those with diabetes and controls (0). The letter following the number corresponds to the parameter(s} assessed in the study: ¢ = extent; i = incidence; p = prevalence; s = severity; r= progression. Karjalainen 1997). All four studies reported greater prevalence, extent, or severity of at least one measure of periodontal disease. Type 2 Diabetes. There are fewer reports on the rela- tionship between type 2 diabetes and periodontal dis- ease, particularly where type 2 diabetes is explicitly identified or discernible from the ages of subjects. Seven studies limited to subjects with type 2 diabetes included a comparison group without diabetes. Two of these studies included only adult subjects (Morton et al. 1995, Novaes et al. 1996); the remaining five were large population-based studies of diabetes and periodontal disease in Pima Indians, a group with the highest known prevalence of type 2 diabetes in the world. The Pima Indian studies included subjects aged 5 years and older (Shlossman et al. 1990) or 15 and older (Emrich et al. 1991, Nelson et al. 1990, ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL Taylor et al. 1998a,b). All seven studies reported greater prevalence, extent, or severity of periodontal disease among subjects with diabetes for at least one measure of periodontal disease. Three of these stud- ies were longitudinal (Nelson et al. 1990, Taylor et al. 1998a,b) and showed that the progression of peri- odontal disease was greater in diabetes patients than in individuals without diabetes. In addition to finding significant differences in various measures of periodontal status between sub- jects with and without type 2 diabetes, a number of these reports also provide estimates of association and risk. Using periodontal attachment loss as the measure, Emrich et al. (1991) estimated that people with type 2 diabetes were 2.8 times more likely to have destructive periodontal disease (odds ratio, 2.8; 95 percent CI, 1.9 to 4.1). When they used radi- ographic bone loss as the measure and controlled for Linkages with General Health other important covariates, the estimate rose to 3.4 (odds ratio, 3.4; 95 percent Cl, 2.3 to 5.2). Nelson et al. (1990) quantified the increased risk of advanced periodontal disease in Pima Indians with and without type 2 diabetes, finding the prevalence of periodontal disease in subjects with diabetes to be 2.6 times greater (95 percent Cl, 1.0 to 6.6) than that of sub- jects without diabetes. Taylor et al. (1996), in anoth- er analysis of data from the Pima Indians, reported that type 2 diabetes was a significant risk factor for more severe alveolar bone loss progression (odds ratio, +.2: 95 percent CI, 1.8 to 9.9), in addition to being a significant risk factor for the prevalence of alveolar bone loss. Studies of Individuals with Type 1 or Type 2 Diabetes. Twelve reports included analyses in which subjects with type 1 and type 2 diabetes were not separated. All of these studies were cross-sectional and included adults; two studies included children or adolescents as well (Benveniste et al. 1967; Wolf 1977). Nine of the 12 studies reported greater prevalence, extent, or severity of periodontal disease among the diabetic subjects for at least one measure or index of peri- odontal disease (Bacic et al. 1988, Belting et al. 1964, Bridges et al. 1996, Finestone and Boorujy 1967, Hove and Stallard 1970, Oliver and Tervonen 1993, Sandler and Stahl 1960, Yavuzyilmaz et al. 1996, Wolf 1977). Hove and Stallard (1970) and Benveniste et al. (1967) did not find significant differences in peri- odontal disease between subjects with and without diabetes. The Hove and Stallard report included 28 subjects with diabetes and 16 without diabetes and may not have had enough statistical power to detect clinical differences, although they were able to detect a significantly higher prevalence of gingival vascular changes in subjects with diabetes. Benveniste et al. (1967) commented that their results may have been influenced by use of relatives without diabetes as the comparison group and that the subjects with diabetes were all under reasonably good control with either insulin or dietary regulation. Both factors may have minimized differences between the groups. Diabetes Type Not Specified. The final set of reports on the association between diabetes and periodontal dis- eases consists of seven cross-sectional studies in which the type of diabetes was not specified and was not easily determined from other information pro- vided. Four of the seven studies included only adults (Dolan et al. 1997, Grossi et al. 1994, Mackenzie and Millard 1963, Tervonen and Knuuttila 1986). In the other three studies, subjects ranged in age from childhood to older adulthood (Albrecht et al. 1988, Campbell 1972, Szpunar et al. 1989). Szpunar et al. (1989) presented analyses of two separate national surveys (the National Health and Nutrition Examination Survey, NHANES I, conducted between 1971 and 1974, and the Hispanic Health and Nutrition Examination Survey, HHANES, conducted between 1982 and 1984). All seven studies found subjects with diabetes to have increased prevalence, extent, and severity of periodontal disease. The statistical significance of the diabetes effect was markedly diminished in the final linear regression model used by Szpunar et al. (1989) in analysis of the NHANES I data. Two of the popu- lation-based surveys, Grossi et al. (1994) and Dolan etal. (1997), provided epidemiologic estimates of the association of diabetes and attachment loss severity with odds ratios of 2.3 (95 percent CI, 1.2 to 4.6) and 1,9 (95 percent CI, 1.3 to 3.0), respectively, while controlling for other covariates. Conclusion. Diabetes is a risk factor for the occur- rence and prevalence of periodontal diseases. Although there is insufficient evidence of a causal association, the findings of greater prevalence, sever- ity, or extent of at least one manifestation of peri- odontal disease in individuals with diabetes is remarkably consistent in the overwhelming majority of studies. Furthermore, there are no studies with superior design features in the literature to refute this assessment. The studies were conducted in distinctly different settings, with subjects from different ethnic populations and of different ages, and with a variety of measures of periodontal status. This inevitable variation in methodology and study populations lim- its the possibility that the same biases apply in all the studies. There is a need for further research using stronger designs that also control for confounding variables such as socioeconomic status. Glycemic Control Several lines of evidence support the plausibility that periodontal infections contribute to problems with glycemic control, thus compromising the health of diabetic patients. It has been reported that the chron- ic release of tumor necrosis factor alpha (TNF-a) and other cytokines such as those associated with peri- odontitis interferes with the action of insulin and leads to metabolic alterations (Hotamisligil et al. 1993, Flier 1993). Other studies have noted relation- ships between insulin resistance and active inflam- matory connective tissue diseases (Hallgren and Lundquist 1983, Svenson et al. 1987), other clinical diseases (Beck-Nielsen 1992, Beisel 1975), acute 114 ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL infection (Drobny et al. 1984, Sammalkorpi 1989), and periodontal disease (Grossi et al. 1999). Grossi and Genco (1998) have proposed a model whereby periodontal infection contributes to hyperglycemia and complicates metabolic control in diabetes. Clinical Studies. The effects of periodontal infection on glycemic control have been investigated in a small number of clinical studies that looked at metabolic control at baseline and following various periodontal treatments (see Table 5.5; Aldridge et al. 1995, Christgau et al. 1998, Grossi et al. 1996, 1997, Miller et al. 1992, Seppala and Ainamo 1994, Seppala et al. 1993, Smith et al. 1996, Westfelt et al. 1996, Williams and Mahan 1960, Wolf 1977). One report is based on an epidemiological cohort study (Taylor et al. 1996). The randomized controlled trials of Grossi et al. (1997) involving populations with type 2 diabetes found that use of the systemic antibiotic doxycycline to treat periodontitis patients with diabetes resulted in a transient (3 to 6 months) improvement in glycemic control. On the other hand, the two con- trolled trials conducted in London by Aldridge et al. (1995) involving patients with type 1 diabetes found no effect. Taken together, these three studies provide inconsistent results and are limited in how well they generalize to broader populations. A small uncon- trolled study of 10 patients by Miller et al. (1992) also reported an improvement in glycemic control of diabetic patients whose periodontal disease was treated with mechanical therapy and systemic doxy- cycline. Five of the above-mentioned studies did not include control groups (Miller et al. 1992, Seppala et al. 1993, Smith et al. 1996, Williams and Mahan 1960, Wolf 1977), and four were not specifically designed to address the relationship between peri- odontal therapy and glycemic control (Grossi et al. 1996, Seppala et al. 1993, Smith et al. 1996, Westfelt et al. 1996), although the data collected allowed the investigators to address the issue. One nonrandom- ized but controlled clinical trial of nonsurgical peri- odontal therapy found no significant influence on medical data for the diabetic patients (Christgau et al. 1998). A clear relationship between improvement in periodontal health and glycemic control has not been shown. The studies seem to suggest that antibiotic treatments may help in glycemic control. A recent longitudinal study indicates inflammation may be a precursor to the onset of type 2 diabetes (Schmidt et al. 1999). Thus periodontal infection may contribute to that inflammation. Conclusion. The body of literature concerning the relationship between periodontal infection and Linkages with General Health impaired glycemic control is varied in the strength, quantity, breadth, and consistency of evidence pre- sented. The preliminary evidence, while encourag- ing, does not support a clear-cut conclusion that treating periodontal infection can contribute to man- agement of glycemic control in type 1 or type 2 dia- betes. As the table indicates, only studies using sys- temic antibiotic treatment affected glycemic control favorably. The results suggest that infections other than periodontitis may be implicated or that inten- sive treatment of periodontal infections with sys- temic antibiotics is necessary to affect glycemic con- trol favorably. Further rigorous controlled studies in diverse populations are warranted. The Oral Infection—Heart Disease and Stroke Connection During the past decade, infectious agents have become recognized as causes of systemic diseases, without fever or other traditional signs of infection. Helicobacter pylori is associated with peptic ulcers and, along with Chlamydia pneumoniae and cytomegalovirus, is now thought to be associated with increased risk for cardiovascular disease as well as malignancies (Wu et al. 2000). Studies investigating the relationship between oral and dental infections and the risk for cardiovascular disease suggest that there is potential for oral microorganisms, such as periodontopathic bacteria, and their effects to be linked with heart disease. Mechanisms of Action Infectious agents are thought to affect the risk of heart disease through several possible mechanisms. Bacteria or viruses originating in tissues such as the lungs or oral mucosa may directly infect blood vessel walls. Such infection may be largely asymptomatic, but may cause local vascular inflammation and injury, which would contribute to the development of lipid-rich plaques and atherosclerosis. Bacteria or viruses may also interact with white blood cells or platelets, both of which integrate into the developing atherosclerotic plaque. Cells of the blood vessel wall and white blood cells and platelets can release prostaglandins (especially PGE,), interleukins (espe- cially IL-1), thromboxane B2 (TBX,), and tumor necrosis factor alpha (TNF-«). Bacterial products in the blood may also stimulate liver production of other pro-inflammatory or pro-coagulant molecules such as C-reactive protein and fibrinogen. Microbes may also stimulate expression of tissue factor, which would activate coagulation. During the process of ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL 115 Linkages with General Health TABLE 5.5 Effects of periodontal disease and its treatment on glycemic control: cdinical and epidemiological evidence Number of subjects Metabolic Study Diabetes a. Treatment (ages) Follow-up Control Evidence Design? Type b. Control (ages) Time Periodontal Therapy Qutcome Effects on Metabolic Control Levelt Aldridge = RCT Type 1 a. 16 (16-40) 2months — Experimental group: oral Glycated Periodontal treatment had no et al. 1995, b. 15 (16-40) hygiene instruction, scaling, hemoglobin, — effect on change in glycated Study 1 adjustment of restoration fructosamine — hemoglobin. margins, and reinforcement after 1 month; control group: no treatment Aldridge = RCT Type 1 a. 12 (20-60) 2months Experimental group: oral Glycated Periodontal treatment had no et al. 1995, b. 10 (20-60) hygiene instruction, scaling hemoglobin —_ effect on change in glycated Study 2 and root planing, extractions, hemoglobin. root canal therapy; control group: no treatment Grossiet RCT Type 2 a. 89 (25-65) 12 months Experimental groups received —Glycated The three groups receiving | al. 1996, b.24 (25-65) either systemic doxycycline or hemoglobin doxycycline and ultrasonic 1997 placebo and ultrasonic bacterial curettage showed bactericidal curettage with significant reductions irrigation using either water, (P < 0.04) in mean glycated chlorhexidine, or povidone- hemoglobin at 3 months. iodine Christgau = Treatment =‘ Type 1 a. 20 (30-66) 2months Scaling/root planing; Glycated No effect on glycated hemoglobin. —‘{l-1 et al.1998 study,non- andtype2 b.20 (30-66) subgingival irrigation with hemoglobin RCT chlorhexidine; oral hygiene instruction; and extractions Westfelt Treatment Type 1 a. 20 (45-65) 5 years Baseline oral hygiene Glycated “The mean value of HbA1c I-1 etal.1996 study,non- andtype2 b.20 (45-65) instruction, scaling and root hemoglobin between baseline and 24 months RCT planing followed by periodic was not significantly different from prophylaxis, oral hygiene that between 24 and 60 months.” instruction, localized sub- gingival plaque removal, and surgery at sites with bleeding on probing and a periodontal probing depth of >5 mm Smith Treatment Type 1 a. 18 (26-57) 2months Scaling and root planing with Glycated Found no statistically or clinically [I-17 et al.1996 — study, non- b.0 ultrasonics and curets; oral hemoglobin _ significant change in glycated RCT hygiene instruction hemoglobin. Tayior Historical Type 2 Notreatmentor 24years Not applicable Glycated Those with severe periodontitis Il-2 et al.1996 prospective control subjects hemoglobin — were ~6 times more likely to have cohort 49 (severe poor glycemic control at follow-up. periodontitis) 56 (less severe periodontitis) Miller Treatment Type 1 a.10(not given) weeks Scaling and root planing, Glycated Found decrease in glycated tll et al.1992 study, non- b.0 systemic doxycycline hemoglobin, © hemoglobin and glycated albumin RCT glycated in patients with improvement in albumin gingival inflammation (P < 0.01). Patients with no improvement in gingival inflammation had either no change or increase in glycated hemoglobin post treatment. coagulation, platelets would become trapped in the growing clot or thrombus. Microthrombus formation is one of the key factors in the development of ather- osclerotic plaques. As atherosclerotic plaques enlarge, "16 the lumen of the coronary blood vessels narrows and the blood supply to the heart muscle becomes reduced. A frank heart attack or myocardial infarc- tion results when a larger part of the coronary artery ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL Linkages with General Health Number of subjects Metabolic Study Diabetes _a. Treatment (ages) Follow-up Control Evidence Designé Type b. Control (ages) Time Periodontal Therapy Outcome Effects on Metabolic Control Level? Seppala ‘Treatment Type 1 a.38 for 1 year; 2years Scaling and root planing, Medical Reported an improvement of the MW et al.1993, study, non- 22 for 2 years‘ periodontal surgery, and history for HBA levels of the PIDD and CIDD Seppala and RCT 26 PIDD-ly (48 + 6) extractions baseline subjects (P < 0.068, t-test). Ainamo 12 CIDD-1y (43 + 5) control status; 1994 16 PIDD-2y glycosylated 6 CIDD-2y hemoglobin b.0 Al and blood glucose for assessing response to treatment Williams Descriptive Not speci- 3-7 Extractions, scaling and Insulin 7 of 9 subjects had “significant” tH and Mahan clinical fied a.9 (20-32) months curettage, gingivectomy, requirement; reduction in insulin requirements. 1960 study b.0 systemic antibiotics diabetes control (not operationally defined) Wolf 1977 Treatment Type 1 and . 8-12 Scaling and home care Blood glucose, Compared 23 subjects with lll study,non- type 2 a.117 (16-60) months instruction, periodontal surgery, 24-h urinary improved oral infections with 23 RT b.0 extractions, endodontic glucose, who had no improvement after treatment, restorations,denture insulindose —_—treatment for oral infection and replacement or repair inflammation. The subjects with 2 RCT = randomized controlled trial. otherwise in assigning the evidence levels. b Levels of evidence are delineated in Table 5.3. Note that because this body of literature is small, this review does not distinguish between “well-designed” studies and < PIDD = poorly controlled insulin-dependent diabetes; CIDD = controlled insulin-dependent diabetes. improved oral inflammation and infection tended to demonstrate diabetes control improvement (P < 0.1). However, Wolf states in discussion,“treatment of periodon- tal inflammation and periapical lesions ... does little to improve the control of diabetes.” lumen becomes occluded. Failing to receive enough blood, the heart muscle dies, resulting in an infarct. Animal Studies Bacteria originating in the oral cavity may also con- tribute to platelet clotting or thrombosis, as proposed by Herzberg et al. (1983, 1994). These investigators have suggested that the association between peri- odontal disease and cardiovascular disease may be due in part to the potential for oral bacteria such as S. sanguis and P. gingivalis to induce platelet aggrega- tion. Platelets aggregate in response to these bacteria as a result of mistaken identity: a protein structure on the surface of certain common strains of S. sanguis and P gingivalis mimics the platelet-interactive regions of collagen molecules (Erickson and Herzberg 1993, Herzberg et al. 1994). Exposure of flowing blood to collagen triggers clotting, the cen- ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL tral event in stopping blood flow. When an experi- mental bacteremia was created with a strain of S. san- guis that carried the collagen-like protein, rabbits showed changes in blood pressure, electrocardio- gram readings, heart rate, and cardiac contractility (Herzberg and Meyer 1996, 1998). Platelets also aggregated in the circulation, resulting in significant declines in platelet counts. From the electrocardio- graphic tracings, rabbit heart muscle also appeared to have suffered ischemic damage. The investigators concluded that oral bacteria carrying the collagen- like protein induced platelet aggregation or clotting in the bloodstream. These clots were of sufficient size to obstruct coronary arteries and produce ischemic heart damage, an early warning sign of a heart attack or an infarction. Because S. sanguis is present in large numbers in dental plaque and is a causative agent in infective endocarditis, it is likely that these bacteria 117 Linkages with General Health have an opportunity to induce platelet clotting dur- ing human bacteremias from oral sources. Bacteria- induced platelet clotting could contribute to microthrombosis during the development of athero- sclerotic plaques and occlusive thrombus formation with occasional myocardial infarction. Population-based Studies Any study investigating the possibility of a unique role for oral pathogens as risk factors for cardiovascu- lar disease, including atherosclerosis and the forma- tion of a blood clot in a coronary artery of the heart, which typically precedes myocardial infarction, must take into consideration such known risk factors as smoking, hypertension, obesity, diabetes, genetic sus- ceptibility, and elevated cholesterol. Genco (1998) and Beck et al. (1998) have recently reviewed studies examining the associations between oral conditions (including periodontitis) and atherosclerosis and coronary heart disease, the latter of which affects 12 million people in the United States and is the leading cause of death. These are summarized in Table 5.6. Of the ten studies cited in the table, six are prospective cohort studies, in which oral health sta- tus was established at the outset (baseline) of the study period and the subjects were followed at peri- odic intervals to a previously defined endpoint, for example, diagnosis of coronary heart disease or an acute myocardial infarction, or death. Beck et al. (1996) combined data from the Veterans Administration Dental Longitudinal Study and its parent longitudinal study, the Normative Aging Study, for a total of 203 cases and 891 noncases, to determine whether periodontal disease, judged by measuring alveolar crestal bone, is a risk factor for cardiovascular disease. After adjusting for age, blood pressure, cholesterol, and body mass index, the investigators found that subjects with periodontal disease were 1.5 times more likely to develop coro- nary heart disease over a 25-year period than controls (odds ratio of 1.5). Similarly, after adjusting for age, smoking, and blood pressure, the investigators found that veterans with periodontal disease were 1.9 times more likely to develop fatal coronary heart disease (odds ratio of 1.9). In a longitudinal study to eliminate the potential confounding effects of smoking, Genco et al. (1997) measured the incidence of periodontal disease and cardiovascular disease in 1,372 American Indians of the Gila River Indian Reservation. Although diabetes is prevalent in this population, cigarette smoking is rare in these individuals (a fact confirmed in this study), so it was considered not to be a risk factor for either cardiovascular disease or periodontal disease in this study. Periodontal disease was measured at baseline, and the incidence of cardiovascular disease was followed over the next 10 years. When the analy- sis was restricted to individuals under age 60, the risk of cardiovascular disease was 2.7 times higher in sub- jects with periodontal disease than in those with lit- tle or no periodontal infection. This association was seen even after adjusting for other risk factors for cardiovascular disease or periodontal disease such as age, sex, cholesterol, weight, high blood pressure, diabetes, and insulin use. The investigators conclud- ed that periodontal disease is an important risk fac- tor for cardiovascular disease for individuals under 60 in this group, second only to the presence of long-term diabetes. Further analysis of death due to cardiovascular disease is needed in this population to complete the study. Mattila and coworkers have conducted both prospective and retrospective studies. A prospective study (Mattila et al. 1995) showed that new episodes of myocardial infarction occurred more frequently in subjects with more extensive “dental” disease. The authors used a measure of dental disease that includ- ed a composite index that assessed caries, periodon- titis, pericoronitis, and periapical lesions. The com- posite index estimates the combined infectious load that contributes to many possible oral infections. After combining the prospective study data with data from an earlier retrospective study (Mattila et al. 1989) and adjusting for age, triglyceride levels, cho- lesterol, C-reactive protein, smoking, social class, diabetes, and hypertension, the investigators found a significant association between dental infections and acute myocardial infarction in men under age 50 (P< 0.01). A more recent study (Mattila et al. 2000) com- pared 85 patients with proven coronary heart disease and 53 matched controls. This case-control study showed that dental indices were higher among coro- nary heart disease patients than controls, but the dif- ferences were not statistically significant. Participants in the study were older, which the authors believed was the most likely reason for the results. In the first National Health and Nutrition Examination Survey, 9,670 subjects were followed for over 14 years. DeStefano et al. (1993) found that there was a 25 per- cent increased risk of cardiovascular disease in indi- viduals with periodontitis compared with those with minimal periodontal disease. The strongest associa- tion was seen in men under 50 (relative risk, 1.7). A limitation of this study, which the authors acknowl- edge, was the lack of baseline data on smoking, a major risk factor for both periodontal and cardiovas- cular disease. Morrison et al. (1999) evaluated a ret- rospective cohort study using participants in the 118 ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL Linkages with General Health TABLE 35.6 Summary of studies assessing the association between oral conditions, atherosclerosis, and coronary heart disease Subjects Association Study (cases/ (odds ratio or Evidence Designa controls) Oral Condition Cardiovascular Outcome Adjustments? relative risk) Levels DeStefano Prospective 1,786/7,974 Russell's periodontal Coronary heart disease . Age, sex, race, education, poverty, 1.72 (1.1-2.68) ll-2 et al. 1993 index (admission to hospital or marital status, cholesterol,BM!, (only formen death) diabetes, smoking under age 50) Mattila Prospective 52/162 Dental index (caries, New acute myocardial Smoking, hypertensian, age, sex, Yes, P< 0.01 II-2 et al. 1995 periodontal disease, infarction or death triglycerides, socioeconomic pulpal infection) status, diabetes, lipids, BMI, previous MI Joshipura Prospective 757/44,119 Periodontal disease New coranary heart Age, BMI, exercise, smoking, 1.67 (1.03-2.71) Il-2 et al. 1996 (self-reported), disease alcohol, vitamin C, family history, tooth loss due to Ml . self-reported periodontal disease Beck Prospective 203/891 —Alveolarcrestal bone New coronary heart Age, BMI, total cholesterol, 1.5 (1.04-2.14) 2 et al. 1996 loss disease, fatal coronary socioeconomic status, DBP.LDL, —1.9 (1.0-3.43) heart disease, stroke smoking, cholesterol 2.8 (1.45-5.48) Genco Prospective 68/1,304 —Alveolarcrestal bone New coronary heart Age, sex, smoking, BMI, diabetes, 2.68 (1.35-5.60) lI-2 et al. 1997 loss disease cholesterol, hypertension Morrison Retrospective 10,000 Severe gingivitis; New coronary heart Age, sex, serum total cholesterol, 1.37 (0.80-2.35) 2 et al. 1999 periodontitis; disease; cerebrovascular smoking, diabetes, hypertension —_for periodontitis edentulousness deaths 1.90 (1.17-3.10) for edentulousness for fatal CHD Joshipura Prospective 42,151 Number of teeth lost Ischemic stroke Age, smoking, obesity, alcohol, = <10 teeth 1.75 I1-2 et al. 1999 exercise, aspirin, family history, (1.03-2.99) profession, hypertension, 11-15 teeth 1.95 hypercholesterolemia (1.07-3.64) 17-24 teeth 1.48 Dental index (caries, (1.02-2.13) Mattila Case-Control 100/102 periodontal disease, Acute myocardial HDL, smoking, C-reactive protein, Yes, P< 0.01 2 et al. 1989 pulpal infections) infarctions hypertension, age, cholesterol, Dental index (caries diabetes, social class Grau Case-Control 166/166 _ periodontal disease, Stroke Diabetes, preexisting vascular Odds ratio 2.6 Ie et al. 1997 periapical infections) disease, socioeconomic status, (1.18-5.70) Dental index (caries, smoking Mattila Case-cantrol 85/53 periodontal disease, New coronary heart Age, sex, smoking, socioeconomic — i-2 et al. 2000 pulpal infections) disease class, hypertension, number of teeth, serum lipid levels the case-control studies, the first number represents the cases, and the second the controls. Levels of evidence are delineated in Table 5.3. aFor the prospective studies, the total number of subjects in the cohort is the sum of the two numbers given, the first number of which represents the subjects followed to the endpoint. For bBMI = body mass index; M! = myocardial infarction; LDL = low-density lipoproteins; HDL = high-density lipoproteins. 1970-72 National Canada Survey. In the younger cohort, those under age 69, they found that gingivi- tis, periodontitis, and edentulousness were related to fatal coronary heart disease in a statistically signifi- cant manner. However, in analyzing those over age 70, none of these dental conditions was associated with fatal heart disease. These results were adjusted for age, sex, serum total cholesterol, smoking status, diabetes status, hypertension status, and province of resi- ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL dence. This pattern of higher risk observed among younger subjects may, to some extent, reflect the rel- ative instability of risk estimates. However, it is also possible that periodontal disease, like other co-mor- bid relative risks for coronary heart disease, general- ly declines with age (Semenciw et al. 1988). Wu et al. (1999) found periodontal disease to be a potential factor for coronary heart disease and stroke based on an analysis of the first National 119 Linkages with General Health Health and Nutrition Examination Survey and its 21- year follow-up. In this analysis, periodontitis was found to be a significant risk factor for cerebrovascu- lar disease, in particular nonhemorrhagic stroke. Compared with no periodontal disease, relative risk (95 percent CI) for incident nonhemorrhagic stroke was 2.11 (1.30 to 3.42) for periodontitis. There was no significant relationship for gingivitis or edentu- lousness, which were 1.24 (0.74 to 2.08) and 1.41 (0.96 to 2.06), respectively. The increased relative risk for total cerebrovascular disease and nonhemor- rhagic stroke was not seen for hemorrhagic stroke. Similar relative risks for total cerebrovascular disease and nonhemorthagic stroke associated with peri- odontitis were seen in white men and women and African Americans. A conclusive statement about a cause-and-effect relationship between periodontitis and the risk of developing cerebrovascular disease, in particular nonhemorrhagic stroke, cannot be made at this time. The consistency of the findings in different racial groups and the strength of the association war- rant further examination of the potentially important association between these two clinical conditions, which are highly prevalent in the adult population. In the largest cohort studied, Joshipura et al. (1996) found that among a group of male health pro- fessionals who were relatively homogeneous socio- economically and who self-reported preexisting peri- odontal disease, those with 10 or fewer teeth were at increased risk of new coronary heart disease, com- pared with those with 25 or more teeth (relative risk, 1.67). These results were adjusted for standard car- diovascular disease risk factors. In a case-control study of 166 patients with acute cerebrovascular disease and 166 age- and sex- matched controls, Grau et al. (1997) found that “poor dental status” was independently associated with cerebrovascular ischemia. These results were based on a subgroup of patients and controls who completed the dental examination. A modified form of the Total Dental Index was used to measure dental status. In an 8-year follow-up of 42,151 male health professionals who were free of cardiovascular disease at baseline, Joshipura et al. (1999) reported that edentulousness was associated with an increased risk of ischemic stroke after adjusting for age, smoking, obesity, alcohol, exercise, aspirin, family history of cardiovascular disease, profession, hypertension, and hypercholesterolemia. Conclusion None of the studies reviewed to date achieves the level of rigor that can unequivocally establish peri- odontitis as an independent risk factor for cardiovas- cular disease or stroke. The methods used to measure or identify periodontal disease ranged widely from self-report, to composite indices that included dental caries experience, to precise measures of periodonti- tis severity. Nevertheless, there were consistent find- ings of increased odds ratios and significant proba- bility (P) values pointing to an association of peri- odontal and other oral infections with an increased risk for cardiovascular disease. Further studies are needed to determine whether periodontal disease alone or in the presence of other oral infections is an independent risk factor for cardiovascular or cere- brovascular disease. Research to elucidate the under- lying pathological mechanisms is also essential. Studies must also clarify the potentially confounding effects of sex, age, socioeconomic level, and race/ ethnicity. Periodontal Disease and Adverse Pregnancy Outcomes Preterm birth and low birth weight are considered the leading perinatal problems in the United States (Gibbs et al. 1992). Although infant mortality rates have decreased substantially over the past genera- tion, the incidence of low birth weight (just under 300,000 cases in 1995) has not shown a comparable decline (Institute of Medicine 1985, USDHHS 1984). Over 60 percent of the mortality of infants without structural or chromosomal congenital defects can be attributed to low birth weight (Shapiro et al. 1980). Mechanisms of Action Periodontal disease may contribute to adverse out- comes of pregnancy as a consequence of a chronic oral inflammatory bacterial infection. Toxins or other products generated by periodontal bacteria in the mother may reach the general circulation, cross the placenta, and harm the fetus. In addition, the response of the maternal immune system to the infection elicits the continued release of inflammato- ry mediators, growth factors, and other potent cytokines, which may directly or indirectly interfere with fetal growth and delivery. Evidence of increased rates of amniotic fluid infection, chorioamnion infection, and histologic chorioamnionitis supports an association between preterm birth, low birth weight, and general infection during pregnancy. It is noteworthy that the largest proportion of such infections occurred during the pregnancies of the most premature births (Hillier et al. 1988, 1995). The biological mechanisms involve bacteria-induced activation of cell-mediated immuni- ty leading to cytokine production and the synthesis 120 ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL and release of prostaglandins, which may trigger preterm labor (Hillier et al. 1988). Elevated levels of prostaglandin as well as cytokines (interleukin-1 (IL- 1), interleukin-6 (IL-6), and tumor necrosis factor alpha (TNF-a)) have been found in the amniotic fluid of patients in preterm labor with amniotic fluid infection (Romero et al. 1993), compared with levels in patients with preterm labor without infection. Animal Models A variety of studies have used the pregnant hamster model. Some investigators have examined the effects of lipopolysaccharide, produced by oral gram-neg- ative pathogens, on cytokine production (Collins et al. 1994a). In other studies, hamsters have been infected with P gingivalis, with or without prior immunization. Collins et al. (1994b) challenged the animals with nondisseminating, low levels of P gingi- valis, introduced in a subcutaneous chamber at a dorsolumbar site. Although the doses were insuffi- cient to induce fever or wasting, the hamster litters of the infected animals showed a significant reduction in fetal weight (24 percent) in comparison with con- trol animals (P < 0.0001). This suppressive effect on fetal weight was accompanied by a proportional intrachamber rise in tumor necrosis factor alpha (TNF-a) and prostaglandin E2 (PGE}) (P < 0.0001). Immunization prior to mating did not provide pro- tection from a challenge during pregnancy, but rather potentiated the effects, indicating the potential strength of a chronic infection. In another series of hamster studies, researchers observed the effects of experimental periodontitis on pregnancy outcomes and amniotic fluid mediators (Offenbacher et al. 1998). The investigators noted a 20 percent decrease in fetal weight (P = 0.002). Periodontal infection in the pregnant hamster also was associated with a significant rise in intra-amniotic PGE, from 3.31 + 1.1 to 13.5 + 4.1 micrograms per milliliter (P = 0.03). These data suggest a link be- tween oral infection and changes in the fetal envi- ronment. Epidemiologic Studies Human case-control studies have demonstrated that mothers of low-birth-weight infants born as a result of either preterm labor or premature rupture of membranes tend to have more severe periodontal dis- ease than mothers with normal-birth-weight infants (Offenbacher et al. 1996, 1998). A case was defined as a mother whose baby weighed less than 2,500 grams and who had one or more of the following fac- tors: gestational age of infant of less than 37 weeks; Linkages with General Health preterm labor, or preterm, premature rupture of membranes. Controls were all normal-birth-weight, full-term infants. In a case-control study of 124 sub- jects, the mean clinical periodontal attachment level for the mothers of low-birth-weight babies was 3.10 + 0.74 (SD) millimeters (mm) per site (93 subjects) versus 2.80 + 0.61 (SD) mm per site (31 subjects) for mothers of normal-weight infants (P = 0.038 for all cases and controls). For a subset of mothers for whom this was the first child, the mean clinical peri- odontal attachment level for those with low-birth- weight babies was 2.98 + 0.84 (SD) mm per site (46 subjects) versus 2.56 + 0.54 (SD) mm per site for controls (20 subjects) at P = 0.041 (Offenbacher et al. 1996). This subset was analyzed separately to avoid the confounding effects of mothers With periodontal disease who had previously given birth to low-birth- weight infants but who later had normal-weight infants. Logistic regression models demonstrated that severe periodontal disease was associated with a sev- enfold increase in risk of low birth weight, after con- trolling for known risk factors such as smoking, race, alcohol use, age, nutrition, and genitourinary tract infection. This study suggests an association between periodontal disease and prematurity. In a subsequent case-control study of 44 sub- jects, additional biochemical and microbial parame- ters of periodontal disease status were studied to assess the relationship of current periodontal status to current pregnancy outcome (Offenbacher et al. 1998). Results indicate that PGE, levels in gingival crevicular fluid were significantly higher in mothers of low-birth-weight infants than in controls (131.4 + 21.8 (SE) versus 62.6 + 10.3 (SE) nanograms per mil- liliter), respectively (at P = 0.02). Furthermore, with- in the group of mothers of low-birth-weight infants there was a significant inverse association between birth weight, gestational age, and gingival crevicular fluid PGE, levels (at P = 0.023 for current births). These data suggest that the level of PGE, in gingival crevicular fluid, serving as a marker of current peri- odontal disease activity, varies inversely with birth weight, that is, the higher the PGE; level, the lower the birth weights. In this study the periodontal dis- ease was more severe in mothers with adverse preg- nancy outcomes, as determined by biochemical and microbial biomarkers, but the difference in clinical attachment levels did not reach statistical signifi- cance (P = 0.11). Studies also have been reported in other coun- tries. In the United Kingdom a preliminary analysis of 167 cases and 323 controls did not show an asso- ciation between periodontal disease and pregnancy ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL 121 Linkages with General Health outcomes (Davenport et al. 1998): however, the investigators did not control for confounding factors. Dasayanake (1998) conducted a matched case-con- trol study with 55 cases in Thailand. Gingivitis was associated with a higher risk of having a growth- restricted infant (odds ratio = 0.3; 95 percent CI, 0.12 to 0.72), controlling for mothers height, prenatal care, dental caries status, and the infant's gender. Smoking was not controlled for in this study. Conclusion As a remote gram-negative infection, periodontal dis- ease may have the potential to affect pregnancy out- come. Not all the obstetric risk factors that result in babies being born too soon and too small have been fully identified (Gibbs et al. 1992, McCormick 1985). Oral infections have been investigated as a potential risk factor for preterm labor or premature rupture of membranes, which are major obstetric antecedents to spontaneous preterm births. Although the findings from animal research and case-control studies are promising, additional work, including longitudinal studies, research on mechanisms, and intervention trials, is needed to determine whether periodontitis is a risk factor and what the mechanisms of action may be for adverse pregnancy outcomes. In the United States, longitudinal and intervention studies are under way. Conclusion This critical Jook at the emerging associations among oral infections and specific conditions establishes the need for an aggressive research agenda to better understand the specific aspects of these associations and the underlying mechanisms. Prospective and intervention studies are under way and should pro- vide additional and stronger evidence of the presence and direction of an association. It is essential for such studies to include populations at known risk for the underlying conditions as well as the general popula- tion. Of the conditions reviewed, the relationship of periodontal disease and diabetes has the strongest evidence, demonstrating that the risk of periodontitis is higher in individuals with diabetes. However, the effect of periodontitis on glycemic control is less clear, a reflection of the difficulty of controlling for the effect of systemic antibiotic treatments used to manage periodontal disease in diabetic patients in clinical trials. IMPLICATIONS OF THE LINKAGES This review of oral health linkages with general health reveals implications for the clinical practice of both medicine and dentistry. The recognition of well- known and established signs and symptoms of oral diseases may assist in the early diagnosis and prompt treatment of some systemic diseases and disorders. The presence of these signs also may lead to the insti- tution of enhanced disease prevention and health promotion procedures. All health professionals, and the public, should be aware of these signs and symp- toms. Individuals, practitioners, and community pro- grams may also benefit from the accelerated develop- ment and testing of readily accessible, acceptable, and simple oral-based diagnostics. A better understanding of the role of the oral cav- ity and its components in protecting against infection is needed. This information should permit the devel- opment of interventions to enhance these compo- nents. For example, research is under way investigat- ing how to augment some of the natural antimicro- bial molecules that are present in saliva and how to use oral and nasal vaccination routes to enhance immunity. Also, host susceptibility factors contribut- ing to the dissemination of oral infections to other parts of the body should be investigated, especially in populations at high risk for disease and infection. In addition, further studies are needed to elucidate the role of the mouth as a means of transmitting infec- tious microbes. This in turn will allow the develop- ment of interventions to prevent transmission and curb the progression of infections once established. The associations between oral infections and dia- betes, heart disease and stroke, and adverse pregnan- cy outcomes warrant a comprehensive and targeted research effort. If any of these associations prove to be causal, major changes in care delivery and in the training of health professionals will be needed. Awareness of the oral complications of medica- tions and other therapies for disease management and for health promotion needs to be enhanced among health care professionals, the public, drug manufacturers, and the research community. For some of these therapies, known interventions exist and should be followed before initiating the therapy to minimize or modulate its side effects. To prevent the oral complications of other therapies, new approaches are needed. Ultimately, and ideally, the side effects of therapies should be considered in the development of new drugs and biologics. 122 ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL FINDINGS e Many systemic diseases and conditions have oral manifestations. These manifestations may be the initial sign of clinical disease and as such serve to inform clinicians and individuals of the need for fur- ther assessment. e = The oral cavity is a portal of entry as well as the site of disease for microbial infections that affect general health status. e The oral cavity and its functions can be adversely affected by many pharmaceuticals and other therapies commonly used in treating systemic conditions. The oral complications of these therapies can compromise patient compliance with treatment. e Individuals such as immunocompromised and hospitalized patients are at greater risk for gen- eral morbidity due to oral infections. e Individuals with diabetes are at greater risk for periodontal diseases. e Animal and population-based studies have demonstrated an association between periodontal diseases and diabetes, cardiovascular disease, stroke, and adverse pregnancy outcomes. Further research is needed to determine the extent to which these asso- ciations are causal or coincidental. REFERENCES Adams D, Davies G. Gingival hyperplasia induced by cyclosporine A. A report of two cases. Br Dent J 1984 Aug;157(3):89-90. Afzal MA, Buchanan J, Dias JA, Cordeiro M, Bentley ML, Shorrock CA, Minor PD. RT-PCR based diagno- sis and molecular characterisation of mumps viruses derived from clinical specimens collected during the 1996 mumps outbreak in Portugal. J Med Virol 1997 Aug;52(4):349-53. Albrecht M, Banoczy J, Tamas G Jr. Dental and oral symptoms of diabetes mellitus. Community Dent Oral Epidemiol 1988; 16(6):378-80. Aldous JA, Aldous SG. Management of oral health for the HIV-infected patient. J Dent Hyg 1991;65:143-5. Aldridge JP, Lester V, Watts TL, Collins A, Viberti G, Wilson RE Single-blind studies of the effects of improved periodontal health on metabolic control in type 1 diabetes mellitus. J Clin Periodontol 1995;22(4):271-5. Andersen WC, Horton HL. Parietal lobe abscess after routine periodontal recall therapy: report of a case. J Periodontol 1990;61:243-7. Andrews M, Farnham S. Brain abscess secondary to dental infection. Gen Dent 1990;38:224-5. Baba TW, Trichel AM, An L, et al. Infection and AIDS in adult macaques after nontraumatic oral exposure to cell-free SIV: Science 1996;272:1486 (abstract). Linkages with General Health Bacic M, Plancak D, Granic M. CPITN assessment of periodontal disease in diabetic patients. J Periodontol 1988 Dec;59(12):816-22. Baker SB, Weinzweig J, Bartlett SF, Whitaker LA. Brain abscess as a complication of orthognathic surgery: diagnosis, management, and pathophysiology. Plast Reconstruct Surg 1999;104;480-2. Barco CT. Prevention of infective endocarditis: a review of the medical and dental literature. J Periodontol 1991 Aug;62(8):510-23. Baron S, Poast J, Richardson CJ, Nguyen D, Cloyd M. Oral transmission of human immunodeficiency virus by infected seminal fluid and milk: a novel mecha- nism. J Infect Dis 2000;181:498-504. Bayliss R, Clarke C, Oakley CM, Somerville W, Whitfield AG, Young SE. The microbiology and pathogenesis of infective endocafditis. Br Heart J 1983 Dec;50(6):513-9. Beck J, Garcia R, Heiss G, Vokonas PS, Offenbacher S. Periodontal disease and cardiovascular disease. J Periodontol 1996 Oct;67(10 Suppl):1123-37. Beck JD, Offenbacher S, Williams R, Gibbs P, Garcia R. Periodontics: a risk factor for coronary heart disease? Ann Periodontol 1998 Jul;3(1):127-41. Beck-Nielsen H. Clinical disorders of insulin resistance. In: Alberti KG, Defronzo RA, Keen H, Zimmet P edi- tors. International textbook of diabetes mellitus. New York: Wiley; 1992. p. 531-68. Beisel WR. Metabolic response to infection. Annu Rev Med 1975;26:9-20. Belting CM, Hiniker JJ, Dummett CO. Influence of dia- betes mellitus on the severity of periodontal disease. J Periodontol 1964;35:476-80. Benveniste R, Bixler D, Conneally PM. Periodontal dis- ease in diabetics. J Periodontol 1967 Jul-Aug;38(4): 271-9. Berger A, Henderson M, Nadoolman W, et al. Oral cap- saicin provides temporary relief for oral mucositis pain secondary to chemotherapy/radiation therapy. J Pain Symptom Manage 1995;10(3):243-8. Bochud PY, Calandra T, Francioli P. Bacteremia due to viridans streptococci in neutropenic patients: a review. Am J Med 1994;97:256-64. Bridges RB, Anderson JW, Saxe SR, Gregory K, Bridges SR. Periodontal status of diabetic and non-diabetic men: effects of smoking, glycemic control, and socioeconomic factors. J Periodontol 1996 Nov; 67(11):1185-92. Campbell MJ. Epidemiology of periodontal disease in the diabetic and the non-diabetic. Aust Dent J 1972 Aug;17(4):274-8. Carl W. Oral complications of local and systemic cancer treatment. Curr Opin Oncol 1995 Jul;7(4):320-4. Centers for Disease Control (CDC). Pneumocystis pneumonia—Los Angeles. MMWR Morb Mortal Wkly Rep 1981a Jun 5;30(21):250-2. ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL 123 Linkages with General Health Centers for Disease Control (CDC). Kaposi’s sarcoma and Pneumocystis pneumonia among homosexual men-——New York City and California. MMWR Morb Mortal Wkly Rep 1981b Jul 3;30(25):305-8. Centers for Disease Control (CDC). Acquired immuno- deficiency syndrome (AIDS): precautions for health care workers and allied professionals. MMWR Morb Mortal Wkly Rep 1983;32:450-2. Centers for Disease Control (CDC). Update: acquired immunodeficiency syndrome—United States. MMWR Morb Mortal Wkly Rep 1985;34:245-8. Centers for Disease Control (CDC). 1993 revised classi- fication system for HIV infection and expanded sur- veillance case definition for AIDS among adolescents and adults. MMWR Morb Mortal Wkly Rep 1992 Dec 18;41(No. RR-17):1-19. Centers for Disease Control (CDC). 1994 revised classi- fication system for human immunodeficiency virus infection in children less than 13 years of age. MMWR Morb Mortal Wkly Rep 1994:43(No. RR- 12). Centers for Disease Control and Prevention (CDC). Transmission of HIV possibly associated with expo- sure of mucous membrane to contaminated blood. MMWR Morb Mortal Wkly Rep 1997;46:620-3. Christersson LA, Albini B, Zambon JJ, Wikesjo UM, Genco RJ. Tissue localization of Actinobacillus actin- omycetemcomitans in human periodontitis. 1. Light, immunofluorescence and electron microscopic stud- ies. J Periodontol 1987a Aug:58(8):529-39. Christersson LA, Wikesja UM, Albini B, Zambon JJ, Genco RJ. Tissue localization of Actinobacillus actin- omycetemcomitans in human periodontitis. II. Correlation between immunofluorescence and cul- ture techniques. J Periodontol 1987b Aug;58(8): 540-5. Christgau M, Palitzsch KD, Schmalz G, Kreiner U, Frenzel S. Healing response to non-surgical peri- odontal therapy in patients with diabetes mellitus: clinical, microbiological, and immunologic results. J Clin Periodontol 1998 Feb;25(2):112-24. Cianciola LJ, Park BH, Bruck E, Mosovich L, Genco RJ. Prevalence of periodontal disease in insulin-depend- ent diabetes mellitus (juvenile diabetes). J Am Dent Assoc 1982 May;104(5):653-60. Coates E, Slade GD, Goss AN, et al. Oral conditions and their social impact among HIV dental patients. Aust Dent J 1996;41:33-6. Cohen DW, Friedman LA, Shapiro J, Kyle GC, Franklin S. Diabetes mellitus and periodontal disease: two- year longitudinal observations. I. J] Periodontol 1970 Dec;41(12):709-12. Collins JG, Smith MA, Amold RR, Offenbacher S. Effects of Escherichia coli and Porphyromonas gingi- valis lipopolysaccharide on pregnancy outcome in the golden hamster. Infect Immun 1994a;62(10): 4652-5. Collins JG, Windley HW 3rd, Arnold RR, Offenbacher S. Effects of a Porphyromonas gingivalis infection on inflammatory mediator response and pregnancy out- come in hamsters. Infect Immun 1994b Oct;62(10): 4356-61. Cone EJ. Saliva testing for drugs of abuse. Ann NY Acad Sci 1993 Sep 20;694:91-127. Cone EJ, Oyler J, Darwin WD. Cocaine disposition in - saliva following intravenous, intranasal, and smoked administration. J Anal Toxicol 1997 Oct;21(6): 465-75. Constantine NT, et al. Diagnostic challenges for rapid human immunodeficiency virus assays: performance using HIV-1 Group O, HIV-1 Group M, and HIV-2 samples. J Hum Virol 1997:145-51. Cruz GD, Lamster IB, Begg MD, et al. The accurate diagnosis of oral lesions in human immunodeficien- cy virus infection. Impact on medical staging. Arch Otolaryngol Head Neck Surg 1996;122:68-73. Dabbs JM Jr. Salivary testosterone measurements in behavioral studies. Ann NY Acad Sci 1993 Sep 20; 694:177-83. Dajani AS, Taubert KA, Wilson W, Bolger AF Bayer A, Ferrieri P Gewitz MH, Shulman ST, Nouri S, Zuccaro G Jr, et al. Prevention of bacterial endocarditis: rec- ommendations by the American Heart Association. Clin Infect Dis 1997 Dec;25(6):1448-58. Dasanayake AP. Poor periodontal health of the pregnant woman as a risk factor for low birth weight. Ann Periodontal 1998;3:206-12. Davenport ES, et al. The East London study of maternal chronic periodontal disease and preterm low birth weight infants: study design and prevalence data. Ann Periodontol 1998;3:213-21. de Pommereau V, Dargent-Pare C, Robert JJ, Brion M. Periodontal status in insulin-dependent diabetic adolescents. J Clin Periodontol 1992 Oct;19(9 Pt.1): 628-32. DeStefano F Anda RF Kahn HS, Williamson DF Russell CM. Dental disease and risk of coronary heart dis- ease and mortality. BMJ 1993 Mar;306(6879): 688-91. Dillon B, Hecht FM, Swanson M, Goupil-Sormany I, Grant RM, Chesney MA, Kahn JO. Primary HIV infections associated with oral transmission (abstract 473). 7th Conference on Retroviruses and Opportunistic Infections; 2000 Jan 30-Feb 2. Chicago, Il. Dodd CL, Greenspan D, Katz MH, Westenhouse JL, Feigal DW, Greenspan JS. Oral candidiasis in HIV infection: pseudomembranous and erythematous candidiasis show similar rates of progression to AIDS. AIDS 1991 Nov;5(11):1339-43. Dolan TA, Gilbert GH, Ringelberg ML, Legler DW, Antonson DE, Foerster U, Heft MW. Behavioral risk indicators of attachment loss in adult Floridians. J Clin Periodontol 1997 Apr;24(4):223-32. Drobny EC, Abramson EC, Baumann G. Insulin recep- tors in acute infection: a study of factors conferring insulin resistance. J Clin Endocrinol Metab 1984 Apr;58(4):710-6. 124 ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL Duffy RE, Adelson R, Niessen LC, Wescott WB, Watkins K, Rhyne RR. VA oral HIV surveillance pro- ram: understanding the disease. J Am Dent Assoc 1992 Oct;123(10):57-62. Durack DT. Prevention of infective endocarditis. N Engl J Med 1995 Jan 5:332(1):38-44. Eisenberg E, Krutchkoff D, Yamase H. Incidental oral hairy leukoplakia in immunocompetent persons. A report of two cases. Oral Surg Oral Med Oral Pathol 1992 Sep;74(3):332-3. Ellison PT. Measurements of salivary progesterone. Ann NY Acad Sci 1993 Sep 20:694:161-76. Emrich LJ, Shlossman M, Genco RJ. Periodontal disease in non-insulin-dependent diabetes mellitus. J Periodontol 1991 Feb;62(2):123-31. Engelhard D, Elishoov H, Or R, et al. Cytosine arabi- noside as a major risk factor for Streptococcus viri- dans septicemia following bone marrow transplanta- tion: a 5-year prospective study. Bone Marrow Transpl 1995;16:565-70. Epstein JB, Silverman S Jr. Head and neck malignancies associated with HIV infection. Oral Surg Oral Med Oral Pathol 1992 Feb;73(2):193-200. Epstein JB, Rea G, Wong FL, Spinelli J, Stevenson- Moore P. Osteonecrosis: study of the relationship of dental extractions in patients receiving radiotherapy. Head Neck Surg 1987;10:48-54. Epstein JB, Priddy RW, Sherlock CH. Hairy leukoplakia- like lesions in immunosuppressed patients following bone marrow transplantation. Transplantation 1988 Sep;46(3):462-4. Erickson PR, Herzberg MC. Evidence for the covalent linkage of carbohydrate polymers to a glycoprotein from Streptococcus sanguis. J Biol Chem 1993 Nov 15;268(32):23780-3. Falk KI, Zou JZ, Lucht E, Linde A, Erberg I. Direct identification by PCR of EBV types and variants in clinical samples. J Med Virol 1997 Apr;51(4):355-63. Faruque S, Edlin B, McCoy CB, et al. Crack cocaine smoking and oral sores in three inner-city neighbor- hoods. } Acquir Immun Defic Syndr Hum Retrovirol 1996a;7-92. Faruque S, Edlin BR, McCoy CB, et al. Oral lesions among crack smokers in three inner-city neighbor- hoods: implications for HIV transmission through oral sex. 3rd Conf Retro Opportun Infect 1996b;73. Faulconbridge AR, Bradshaw WC, Jenkins PA, Baum JD. The dental status of a group of diabetic children. Br Dent J 1981 Oct 20;151(8):253-5. Feigal DW, Katz MH, Greenspan D, Westenhouse J, Winkelstein W Jr, Lang W, Samuel M, Buchbinder SP, Hessol NA, Lifson AR, et al. The prevalence of oral lesions in HIV-infected homosexual and bisexual men: three San Francisco epidemiological cohorts. AIDS 1991 May;5(5):519-25. Feld R. The role of surveillance cultures in patients like- ly to develop chemotherapy-induced mucositis. Support Care Cancer 1997 Sep;5(5):371-5. Linkages with General Health Felix DH, Watret K, Wray D, Southam JC. Hairy leuko- plakia in an HIV-negative, nonimmunosuppressed patient. Oral Surg Oral Med Oral Pathol 1992 Nov;74(5):563-6. Field MJ, Lawrence RL, Zwanziger L, editors. Extending Medicare coverage for preventive and other services. Washington: Institute of Medicine, National Academy Press, 2000. Finestone AJ, Boorujy SR. Diabetes mellitus and peri- odontal disease. Diabetes 1967 May;16(5):336-40. Firatli E. The relationship between clinical periodontal status and insulin-dependent diabetes mellitus. Results after 5 years. J Periodontol 1997 Feb;68(2): 136-40. Firatli E, Yilmaz O, Onan U. The relationship between clinical attachment loss and the duration of insulin- dependent diabetes mellitus (IDDM) in children and adolescents. J Clin Periodontol 1996 Apr;23(4): 362-6. Fisch! MA, Dickinson GM, Scott GB, et al. Evaluation of heterosexual partners, children, and household con- tacts of adults with AIDS. JAMA 1987;257:640-4. Flier JS. An overview of insulin resistance. In: Moller DE, editor. Insulin resistance. New York: Wiley; 1993. p. 1-8. Fourrier E Duvivier B, Boutigny H, Roussel-Devallez M, Chopin C. Colonization of dental plaque: a source of nosocomial infections in intensive care unit patients. Crit Care Med 1998;26:301-8. Galea H, Aganovic I, Aganovic M. The dental caries and periodontal disease experience of patients with early onset insulin dependent diabetes. Int Dent J 1986 Dec;36(4):219-24. Gallagher DM, Erickson Kk, Hollin SA. Fatal brain abscess following periodontal therapy: a case report. Mt Sinai J Med 1981:48:158-60. Gandolfo AM, De LP, Modugno G, et al. The oral status of HIV+ patients. The clinico-statistical results with reference to 2 examination check-ups. [Ital]. Minerva Stomatologica 1991:40:651-6. Genco RJ. Periodontal disease and risk for myocardial infarction and cardiovascular disease. Cardiovasc Rev Rep 1998;19(3):34-40. Genco R, Chadda S, Grossi S, Dunford R, Taylor G, Knowler W, Pettitt D. Periodontal disease is a predic- tor of cardiovascular disease in a Native American population. J Dent Res 1997 Mar;76(Spec No):14- 519 (abstract 3158). Geraci JE, Wilson JR. Symposium on infective endo- carditis. Ill. Endocarditis due to gram-negative bac- teria. Report of 56 cases. Mayo Clin Proc 1982 Mar;57(3):145-8. Gibbs RS, Romero R, Hillier SL, Eschenbach DA, Sweet RL. A review of premature birth and subclinical infection. Am J Obstet Gynecol 1992 May;166(5): 1515-28. ORAL HEALTH IN AMERICA: A REPORT OF THE SURGEON GENERAL 125