The association between diabetes mellitus and periodontal disease has long been discussed, with conflicting conclusions. On the one hand, numerous reports indicate a high prevalence of periodontal disease in diabetics compared to healthy controls, while others fail to show such a relationship. Clarification of this dilemma has been occurring as the diagnostic criteria for periodontal disease destruction improve and the number and size of the populations surveyed grow. This review is based on a selective review of the literature from the present decade. To date, based mainly on an extensive study of the Pima Indians who have an extremely high incidence of non-insulin-dependent diabetes mellitus (NIDDM), it seems to be clear that patients with NIDDM have a higher prevalence and severity of periodontal disease destruction than non-diabetics in the same population. However, it must be borne in mind that these data are for a special population. Studies on patients with insulin-dependent diabetes mellitus (IDDM) indicate results similar to those found in studies on NIDDM. There is an increase in prevalence and severity of periodontitis compared to controls. For both IDDM and NIDDM, there does not appear to be any correlation between the prevalence or the severity of periodontal disease and the duration of diabetes. Well-controlled diabetic patients as measured by blood glycated hemoglobin levels have less severe periodontal disease than poorly controlled diabetics. The principles of treatment of periodontitis in diabetics are the same as those for non-diabetic patients and are consistent with our approach to all high-risk patients who have already developed periodontal disease. The major efforts should be directed at the prevention of periodontitis in patients at risk of developing diabetes. Another important clinical question relates to the influence of periodontal disease on the control of the diabetic state. Here again the literature is unclear; however, a recent development suggests that effective control of periodontal infection in patients with diabetes reduces the level of advanced glycosylation end products in the serum. If future studies can confirm this effect, then periodontal infection control must be considered an integral part of diabetic control.