Associations Between Periodontal Disease and ... - Semantic Scholar

J Periodontol • November 2005 (Suppl.)

Associations Between Periodontal Disease and Systemic Disease: Evaluating the Strength of the Evidence Thomas Dietrich*† and Raul I. Garcia*‡

Much work has been published on the association between periodontal disease and systemic disease, including original reports, narrative reviews, systematic reviews, and metaanalyses. Based on the existent work, one can assign an evidence level and grade, using standard evidence-based criteria, to the data available in the four major categories of medical outcomes studied: cardiovascular/cerebrovascular, pregnancy, pulmonary, and diabetes. We discuss methodologic and conceptual problems in the study of oral-systemic associations, focusing as an example on the association between periodontal disease and cardiovascular/cerebrovascular disease. We argue that the hierarchical ranking of studies by levels of evidence may be misleading. In particular, while randomized controlled trials (RCTs) are needed to determine the efficacy of periodontal treatment to reduce the risk of cardiovascular events, they may be of limited value in determining the etiologic role of periodontal disease on coronary heart disease and stroke. We discuss limitations of RCTs as well as the limitations of currently available data from epidemiologic studies, including study design and confounding and misclassification errors. We conclude that well-designed observational studies into the associations between periodontal disease and systemic disease need to remain an integral component of future research efforts in order to fully understand such associations. J Periodontol 2005;76:2175-2184. KEY WORDS Cerebrovascular disease; coronary heart disease; epidemiologic studies; periodontal disease; randomized controlled trials; stroke.

* Department of Health Policy and Health Services Research, Boston University Goldman School of Dental Medicine, Boston, MA. † Department of Periodontology and Oral Biology, Boston University Goldman School of Dental Medicine. ‡ VA Normative Aging Study, VA Boston Healthcare System.

T

he possible role of periodontal infections in the pathogenesis of systemic diseases remains an important but unresolved question. These symposium presentations highlight the breadth and depth of interest on this topic. In less than a decade, there has been extraordinary growth in the evidence base, including laboratory, animal, and human studies, on the relation between periodontal disease and important systemic health outcomes. Furthermore, a number of systematic reviews and meta-analyses on specific oralsystemic associations have been published.1-7 We will focus here on some methodological issues that are important in evaluating clinical epidemiologic research and in guiding plans for randomized controlled trials. We will focus on the association between periodontal disease and coronary heart disease (CHD) and ischemic stroke. However, many of the points made apply to the other systemic health outcomes under study. CAUSAL THINKING VERSUS LEVELS OF EVIDENCE A key question is whether periodontitis is a causal factor in the etiology of CHD and/or stroke. Much epidemiologic research is devoted to the estimation of causal effects using non-experimental data. Unfortunately, the results of epidemiologic studies, typically yielding a measure of association, are insufficient to conclude whether such associations are causal. The correct interpretation of 2175

Periodontal Disease and Systemic Disease

statistical associations between periodontal disease and CHD depend on a variety of issues, including but not limited to, confounding bias, other sources of bias, and chance. Several criteria have been suggested for evaluating the likelihood that exposure–disease8 associations in observational studies are causal. Such criteria, in particular the Bradford Hill criteria, have been previously applied and discussed in the context of the periodontal disease–cardiovascular disease association.9,10 It is well recognized that the application of these criteria, while suitable to guide one’s thinking, will not provide a definitive answer. The limitations of these criteria are many and have led some to propose the term causal ‘‘values’’ rather than causal ‘‘criteria.’’11,12 Hence, conclusions reached by weighting the available evidence by different researchers may differ. In an attempt to evaluate and quantify this ‘‘conclusion-making’’ process, Holman et al.13 conducted a psychometric experiment on causal inference among 159 epidemiologists who were asked to decide whether to attribute causality to 12 summaries of evidence concerning a disease and a chemical exposure. Factors with the strongest influence on the odds of causal attribution were statistical significance; refutation of alternative explanations; strength of association; and adjunct information concerning biological, factual, and theoretical coherence; and the number of supportive studies.13 The results demonstrate that in epidemiologic practice, conclusions are reached by weighing evidence from several sources and not merely by combining the quantitative results of clinical studies. Causal criteria are inevitably applied by epidemiologists and clinical researchers, formally or informally, on an everyday basis. However, in the absence of experimental evidence from randomized controlled trials (RCTs), the determination of an observed association as causal is neither truly objective nor easy. As Kaufman and Poole noted, ‘‘no final resolution has emerged to the challenges of thinking about causes and their effects or to the formidable task of forming causal judgments about relations between variables.’’11 These concerns also apply directly to the current status of work on periodontal-systemic associations. In the era of evidence-based medicine (EBM), it is only natural for a clinical scientist to feel somewhat uncomfortable with such subjective criteria. With the advent of EBM, a ranking system based on levels of evidence has been developed (Table 1). It is important to note that the concept of EBM was originally developed to evaluate interventions, rather than to address etiologic research questions per se.14,15 However, the Oxford Centre for Evidence-Based Medicine Levels of Evidence now specifically include ‘‘Etiology’’ as a category.16 2176

Volume 76 • Number 11 (Suppl.)

Table 1.

Levels of Evidence for Etiologic Studies as Proposed by Oxford Centre of Evidence-Based Medicine16 Level 1

Description a b c

2

a b c

3

a b

SR with homogeneity of RCT Individual RCT with narrow confidence interval All or none SR with homogeneity of cohort studies Individual cohort study (including low quality RCT; e.g., 6 mm the OR was 4.3 (95% CI:1.8 to 10.2). Furthermore, results of stratified analyses suggested that the association was stronger in younger subjects and in males (Fig. 1). Hence, the study supports the hypothesis that the association between periodontal disease and cardiovascular disease may be stronger in younger subjects, as previously reported,3 and suggests that the association may be stronger in males. However, the study was not designed to evaluate such effect modification and the power of these subgroup analyses is limited. A large multicenter case-control study would allow evaluating the association in different subgroups in an efficient manner. Such findings may not only contribute to our understanding of the periodontal disease–cardiovascular disease association, but also have important implications in regards to the design of intervention studies. For example, if the association between periodontitis and stroke was, in fact, very small or absent in older subjects, the rationale for an intervention study for the secondary prevention of cerebrovascular events in older subjects would have to be questioned. CONFOUNDING Confounding is an important issue that invariably arises when causal effects are estimated from nonexperimental data.51 In the case of the periodontal disease–cardiovascular disease association, confounding arises from risk factors for cardiovascular disease that are independently associated with peri-

Figure 1. The association between history of periodontitis and acute cerebral ischemia; results from a case-control study.27 Adjusted OR for all subjects (red squares) by mean CAL and stratified by age (blue diamonds) and gender (green triangles).

Dietrich, Garcia

odontal disease. The amount of bias introduced by a confounding variable depends on the strength of the association with both periodontal disease and cardiovascular disease as well as the prevalence of the confounding variable in the study population. Smoking is an established risk factor for CHD and stroke.52 It is also an established risk factor for periodontal disease.53 Furthermore, cigarette smoking is a relatively common exposure. Therefore, smoking is a very important confounding variable in the association of periodontal disease and cardiovascular disease. Even if smoking is adjusted for in the statistical analysis, residual confounding by smoking history is of great concern, particularly since the reported associations between periodontal disease and CHD/stroke are small. The role of smoking as a confounder of the oral health–systemic disease association is receiving increasing attention.41,54 One possible source of residual confounding by smoking is the measurement error associated with self-reported smoking measures.41,55 However, even if smoking history could be accurately measured, modeling smoking history in statistical models is not straightforward, as smoking is a multidimensional phenomenon with various characteristics like intensity, duration and time since cessation.56,57 Hence, another possible source of residual confounding is the categorization of continuous smoking characteristics like intensity or duration.58-60 These inherent difficulties in assessing all dimensions of cigarette smoking and the inability to completely account for them in statistical models have led some authors to suggest the restriction of studies on periodontal disease–CHD/stroke associations to never smokers to avoid residual confounding.54 However, this goal may be difficult to achieve for practical reasons. More importantly, effect modification by smoking is an alternative explanation if an exposure– disease association can only be demonstrated in smokers and results found among never-smokers may not be generalizable to smokers.61,62 It may be difficult to decide between residual confounding and effect modification in specific analyses.6 Unfortunately, the currently available biomarkers are of limited value for assessment of smoking history in epidemiologic studies on the association between periodontal disease and CHD/stroke.63 Biomarkers such as cotinine can confirm that a subject is currently non-smoking and correlate with smoking intensity in current smokers. However, they do not provide information on smoking duration nor are they useful to assess smoking history in former smokers. In countries where the smoking epidemic passed its peak years ago, such as the United States, the proportion of former smokers among ever smokers has risen constantly and former smokers constitute an important 2181


population in clinical research. This is particularly relevant in studies of older adults, where, in addition to the overall decline in smoking, the increased mortality among smokers contributes to the higher proportion of former smokers. This age-dependent rise in the proportion of former smokers among ever smokers has been demonstrated in a series of representative surveys between 1980 and 1992.64 Hence, effect estimates should be reported separately for never smokers, which will convey important additional information. However, statistical methods that make efficient use of the information collected on smoking history are desirable. We have recently proposed a comprehensive smoking index that simultaneously accounts for intensity, duration, and time since cessation for former smokers.65 This index can also account for changes in smoking intensity over time which is usually ignored in analyses. Such approaches may also be useful to reduce residual confounding by smoking history. Importantly, in their meta-analysis of available cohort studies, Janket et al. identified residual confounding as a source of heterogeneity between studies.3 Hence, residual confounding may explain the small associations found in the available cohort studies. CONCLUSIONS The currently available evidence from epidemiologic studies suggests that there is a significant, although modest, association between periodontal disease and coronary heart disease and other systemic health outcomes. However, there is concern that this association may be due to confounding by smoking or other, unknown factors. In contrast, misclassification of the periodontal exposure may attenuate the strength of the observed relationship. The currently available data from observational studies are insufficient to accurately estimate the strength of the association. RCTs to evaluate the effect of periodontal therapy in patients with chronic periodontitis on inflammatory biomarkers, other cardiovascular risk factors, and ultimately on risk of cardiovascular events, are desirable and important. However, such RCTs may not yield a definitive answer as to whether periodontal disease is causally related to coronary heart disease/ stroke or any other systemic health outcome. There will remain a need for well-designed, carefully analyzed, and correctly interpreted observational studies, including case-control-studies, in order to adequately understand the association between periodontal disease and systemic disease in various populations. ACKNOWLEDGMENTS This study was supported by the United States Department of Veterans Affairs (Epidemiology Merit Review 2182


Program and CSP-ERIC Program) and by NIDCR grant K24 DE00419. REFERENCES 1. Scannapieco FA, Bush RB, Paju S. Associations between periodontal disease and risk for atherosclerosis, cardiovascular disease, and stroke. A systematic review. Ann Periodontol 2003;8:38-53. 2. Scannapieco FA, Bush RB, Paju S. Periodontal disease as a risk factor for adverse pregnancy outcomes. A systematic review. Ann Periodontol 2003;8:70-78. 3. Janket SJ, Baird AE, Chuang SK, Jones JA. Metaanalysis of periodontal disease and risk of coronary heart disease and stroke. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2003;95:559-569. 4. Madianos PN, Bobetsis GA, Kinane DF. Is periodontitis associated with an increased risk of coronary heart disease and preterm and/or low birth weight births? J Clin Periodontol 2002;29(Suppl. 3):22-36; discussion 37-38. 5. Khader YS, Albashaireh ZS, Alomari MA. Periodontal diseases and the risk of coronary heart and cerebrovascular diseases: A meta-analysis. J Periodontol 2004;75:1046-1053. 6. Garcia RI, Nunn ME, Vokonas PS. Epidemiologic associations between periodontal disease and chronic obstructive pulmonary disease. Ann Periodontol 2001; 6:71-77. 7. Hujoel PP. Does chronic periodontitis cause coronary heart disease? A review of the literature. J Am Dent Assoc 2002;133(Suppl.):31S-36S. 8. Hill AB. The environment and disease: Association or causation? Proc R Soc Med 1965;58:295-300. 9. Beck J, Garcia R, Heiss G, Vokonas PS, Offenbacher S. Periodontal disease and cardiovascular disease. J Periodontol 1996;67(Suppl.):1123-1137. 10. Meurman JH, Sanz M, Janket SJ. Oral health, atherosclerosis, and cardiovascular disease. Crit Rev Oral Biol Med 2004;15:403-413. 11. Kaufman JS, Poole C. Looking back on ‘‘causal thinking in the health sciences.’’ Annu Rev Public Health 2000;21:101-119. 12. Poole C. Causal values. Epidemiology 2001;12:139141. 13. Holman CD, Arnold-Reed DE, de Klerk N, McComb C, English DR. A psychometric experiment in causal inference to estimate evidential weights used by epidemiologists. Epidemiology 2001;12: 246-255. 14. Sackett DL. Rules of evidence and clinical recommendations on the use of antithrombotic agents. Chest 1986;89:2S-3S. 15. Cook DJ, Guyatt GH, Laupacis A, Sackett DL, Goldberg RJ. Clinical recommendations using levels of evidence for antithrombotic agents. Chest 1995;108:227S230S. 16. Levels of Evidence and Grades of Recommendation. Oxford Centre for Evidence Based Medicine 2004. Available at http://www.cebm.net/levels_of_evidence. asp; accessed October 20, 2004. 17. Scannapieco FA, Bush RB, Paju S. Associations between periodontal disease and risk for nosocomial bacterial pneumonia and chronic obstructive pulmonary disease. A systematic review. Ann Periodontol 2003;8:54-69.

J Periodontol • November 2005 (Suppl.)

18. Niederman R, Joshipura KJ. Cause celebre: Oral health and heart disease. Evid Based Med 2000; 2:59-60. 19. Hernan MA. A definition of causal effect for epidemiological research. J Epidemiol Community Health 2004;58:265-271. 20. Rich SE, Wehler CJ, McCoy LC, Miller DR, Rothendler JA, Jones JA. Withdrawals in the VA Dental Diabetes Study (VA-DDS). J Dent Res 2005;84(Spec. Issue A): 3149. 21. Grayston JT, Kronmal RA, Jackson LA, et al. Azithromycin for the secondary prevention of coronary events. N Engl J Med 2005;352:1637-1645. 22. Cannon CP, Braunwald E, McCabe CH, et al. Antibiotic treatment of Chlamydia pneumoniae after acute coronary syndrome. N Engl J Med 2005;352:1646-1654. 23. Rothman KJ, Greenland S. Case-control studies. In: Rothman KJ, Greenland S, eds. Modern Epidemiology. Philadelphia: Lippincott Williams & Wilkins; 1998: 93-114. 24. Wacholder S. Design issues in case-control studies. Stat Methods Med Res 1995;4:293-309. 25. Wacholder S, McLaughlin JK, Silverman DT, Mandel JS. Selection of controls in case-control studies. I. Principles. Am J Epidemiol 1992;135:1019-1028. 26. Rothman KJ, Greenland S. Types of epidemiologic studies. In: Rothman KJ, Greenland S, eds. Modern Epidemiology. Philadelphia: Lippincott Williams & Wilkins; 1998:67-78. 27. Grau AJ, Becher H, Ziegler CM, et al. Periodontal disease as a risk factor for ischemic stroke. Stroke 2004;35:496-501. 28. Dorfer CE, Becher H, Ziegler CM, et al. The association of gingivitis and periodontitis with ischemic stroke. J Clin Periodontol 2004;31:396-401. 29. Hujoel PP, Drangsholt M, Spiekerman C, DeRouen TA. Periodontal disease and coronary heart disease risk. JAMA 2000;284:1406-1410. 30. DeStefano F, Anda RF, Kahn HS, Williamson DF, Russell CM. Dental disease and risk of coronary heart disease and mortality. Br Med J 1993;306:688691. 31. Wu T, Trevisan M, Genco RJ, Dorn JP, Falkner KL, Sempos CT. Periodontal disease and risk of cerebrovascular disease: The first National Health and Nutrition Examination Survey and its follow-up study. Arch Intern Med 2000;160:2749-2755. 32. Morrison HI, Ellison LF, Taylor GW. Periodontal disease and risk of fatal coronary heart and cerebrovascular diseases. J Cardiovasc Risk 1999;6:7-11. 33. Joshipura KJ, Rimm EB, Douglass CW, Trichopoulos D, Ascherio A, Willett WC. Poor oral health and coronary heart disease. J Dent Res 1996;75:16311636. 34. Howell TH, Ridker PM, Ajani UA, Hennekens CH, Christen WG. Periodontal disease and risk of subsequent cardiovascular disease in U.S. male physicians. J Am Coll Cardiol 2001;37:445-450. 35. Genco R, Chadda S, Grossi S. Periodontal disease is a predictor of cardiovascular disease in a native American population. J Dent Res 1997;76:408. 36. Tuominen R, Reunanen A, Paunio M, Paunio I, Aromaa A. Oral health indicators poorly predict coronary heart disease deaths. J Dent Res 2003;82:713-718. 37. Joshipura KJ, Hung HC, Rimm EB, Willett WC, Ascherio A. Periodontal disease, tooth loss, and incidence of ischemic stroke. Stroke 2003;34:47-52.

Dietrich, Garcia

38. Flegal KM, Brownie C, Haas JD. The effects of exposure misclassification on estimates of relative risk. Am J Epidemiol 1986;123:736-751. 39. de Klerk NH, English DR, Armstrong BK. A review of the effects of random measurement error on relative risk estimates in epidemiological studies. Int J Epidemiol 1989;18:705-712. 40. Greenland S. Basic methods for sensitivity analysis and external adjustment. In: Rothman KJ, Greenland S, eds. Modern Epidemiology. Philadelphia: Lippincott Williams & Wilkins; 1998:343-358. 41. Spiekerman CF, Hujoel PP, DeRouen TA. Bias induced by self-reported smoking on periodontitis-systemic disease associations. J Dent Res 2003;82:345-349. 42. Hunt RJ, Fann SJ. Effect of examining half the teeth in a partial periodontal recording of older adults. J Dent Res 1991;70:1380-1385. 43. Kingman A, Albandar JM. Methodological aspects of epidemiological studies of periodontal diseases. Periodontol 2000 2002;29:11-30. 44. Joshipura KJ, Pitiphat W, Douglass CW. Validation of self-reported periodontal measures among health professionals. J Public Health Dent 2002;62:115-121. 45. Russell AL. A system of classification and scoring for prevalence surveys of periodontal disease. J Dent Res 1956;35:350-359. 46. Phillips CV. Quantifying and reporting uncertainty from systematic errors. Epidemiology 2003;14:459-466. 47. Lash TL, Fink AK. Semi-automated sensitivity analysis to assess systematic errors in observational data. Epidemiology 2003;14:451-458. 48. Greenland S. The effect of misclassification in the presence of covariates. Am J Epidemiol 1980;112: 564-569. 49. Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology 1990;1:43-46. 50. Savitz DA, Olshan AF. Multiple comparisons and related issues in the interpretation of epidemiologic data. Am J Epidemiol 1995;142:904-908. 51. Greenland S, Morgenstern H. Confounding in health research. Annu Rev Public Health 2001;22:189-212. 52. Burns DM. Epidemiology of smoking-induced cardiovascular disease. Prog Cardiovasc Dis 2003;46:11-29. 53. Tomar SL, Asma S. Smoking-attributable periodontitis in the United States: Findings from NHANES III. J Periodontol 2000;71:743-751. 54. Hujoel PP, Drangsholt M, Spiekerman C, DeRouen TA. Periodontitis-systemic disease associations in the presence of smoking — causal or coincidental? Periodontol 2000 2002;30:51-60. 55. Patrick DL, Cheadle A, Thompson DC, Diehr P, Koepsell T, Kinne S. The validity of self-reported smoking: A review and meta-analysis. Am J Public Health 1994;84:1086-1093. 56. McKnight B, Cook LS, Weiss NS. Logistic regression analysis for more than one characteristic of exposure. Am J Epidemiol 1999;149:984-992. 57. Leffondre K, Abrahamowicz M, Siemiatycki J, Rachet B. Modeling smoking history: A comparison of different approaches. Am J Epidemiol 2002;156:813-823. 58. Cochran WG. The effectiveness of adjustment by subclassification in removing bias in observational studies. Biometrics 1968;24:295-313. 59. Becher H. The concept of residual confounding in regression models and some applications. Stat Med 1992;11:1747-1758. 2183


60. Brenner H, Blettner M. Controlling for continuous confounders in epidemiologic research. Epidemiology 1997;8:429-434. 61. Hyman JJ, Winn DM, Reid BC. The role of cigarette smoking in the association between periodontal disease and coronary heart disease. J Periodontol 2002; 73:988-994. 62. Hyman JJ, Reid BC. Cigarette smoking, periodontal disease and chronic obstructive pulmonary disease. J Periodontol 2004;75:9-15. 63. Scott DA, Palmer RM, Stapleton JA. Validation of smoking status in clinical research into inflammatory periodontal disease. J Clin Periodontol 2001;28:715722.

2184


64. Arnett DK, Sprafka JM, McGovern PG, et al. Trends in cigarette smoking: The Minnesota Heart Survey, 1980 through 1992. Am J Public Health 1998;88:12301233. 65. Dietrich T, Hoffmann K. A comprehensive index for the modeling of smoking history in periodontal research. J Dent Res 2004;83:859-863. Correspondence: Dr. Raul I. Garcia, Department of Health Policy and Health Services Research, Boston University Goldman School of Dental Medicine, 715 Albany St., 560 – 3rd Fl., Boston, MA 02118. Fax: 617/638-6381; e-mail: [email protected]. Accepted for publication January 26, 2005.