Probing pocket depth and attachment loss in NIDDM

Periodontal Disease Progression in Type II Non-Insulin-Dependent Diabetes Mellitus Patients (NIDDM). Part I - Probing Pocket Depth and Clinical Attachment

Arthur B. Novaes Jr.1 Ferney G. Gutierrez2 Arthur B. Novaes3 1Departamento de Periodontologia, Universidade Federal de Rio de Janeiro, Rio de Janeiro, RJ, Brasil 2Department of Periodontology, Metropolitan School of Dentistry, Barranquilla, Colombia 3Departamento de Periodontologia, Faculdade de Odontologia de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto, SP, Brasil

Braz Dent J (1996) 7(2): 65-73 ISSN 0103-6440

Periodontal disease progression of 30 type II diabetic patients (NIDDM) and 30 patients in whom diabetes was not detected was evaluated. Age ranged from 30 to 77 years. To determine the periodontal condition, probing pocket depth and periodontal attachment loss were measured; to determine the metabolic control of the patients, glycosylated hemoglobin and fasting glucose were measured. At the end of the study, the diabetic group was divided into three subgroups, according to the metabolic state of the patients: controlled patients, moderately controlled patients, and poorly controlled patients. Comparing the diabetic and the control groups as a whole, there was no statistically significant difference in probing pocket depth, but significance (P<0.01) was observed for attachment loss. When diabetic patients were divided into subgroups, significant differences were observed between the poorly controlled and the control groups (P<0.01) for both the probing pocket depth and periodontal attachment. The glycosylated hemoglobin test was more reliable than the fasting glucose analysis.

Key Words: Periodontal disease progression; type II diabetes; probing pocket depth; attachment loss.

Introduction

The association between periodontal disease and the presence of microorganisms on dental surfaces is well known and this association as well as the host response are related to the progression of the disease. Frequently, dentists treat patients with systemic problems such as diabetes which can affect the course and therapy of periodontal disease. According to Shlossman (1990), besides the primary etiological factor, systemic diseases are considered to be contributing factors during the initial phase of periodontitis. Literature specific to this subject shows that controversy exists when periodontal disease and diabetes are correlated (Barnett et al., 1984). In order to obtain more accurate studies, instead of considering the patients as one group, it is advisable to separate type I insulin-dependent diabetes mellitus (IDDM) and type II non-insulin-dependent diabetes mellitus (NIDDM) patients and also consider the control of the diabetes. Proof of the interrelationship between diabetes and periodontal disease is difficult; however, it has already been reported that patients with poorly controlled diabetes present a higher incidence of periodontal disease than well-controlled patients. The non-controlled diabetic patient presents problems, not only concerning the control of the periodontal disease but also during clinical treatment (Rhodus, 1987). Goodson et al. (1982), Lindhe et al. (1983), and Harley et al. (1987) report a dynamic pattern for the progression of periodontal disease with periods of exacerbation and remission. However, whether the periodontal disease would follow a similar course in the presence of systemic diseases has not yet been evaluated. The aim of this study was to evaluate whether type II non-insulin-dependent diabetes mellitus (NIDDM) changes the pattern of evolution of periodontal disease using monthly examinations for one year. The following parameters were studied: probing pocket depth (PPD), clinical attachment loss (CAL), and metabolic condition of the patients, using glycosylated hemoglobin and fasting glucose tests.

Material and Methods

The longitudinal method of study was used in order to achieve the proposed objectives of this study. A total of 30 non-diabetic patients (control) and 30 NIDDM patients were examined monthly for one year. The diabetic patients were selected from the Department of Nutrology of the University Hospital and the control group of non-diabetic patients from the Periodontal Clinic of the School of Dentistry of the Federal University of Rio de Janeiro. The patients were selected based on the following criteria: a) presence of periodontal alterations, with a diagnosis of adult periodontitis; b) no periodontal treatment for at least one year before the initial examination and during the study; c) no antibiotic administration during the six months preceding the first examination; d) no family history of diabetes in the control patients. Patients with an initial diagnosis of localized juvenile periodontitis or early onset periodontitis were excluded from the sample. The patients received no periodontal treatment during the study but after the end of the study both diabetic and control groups were treated. The patients agreed to the proposed conditions. The average age of the NIDDM patients in this study was 52.3 years, ranging from 30 to 77 years. The average age of the control patients was 44.6 years, ranging from 30 to 67 years. Examinations were carried out by one of the authors (F.G.G.) lessening the possibility of deviations in the results, such as due to different probing pressures (Listgarten, 1980). PPD and CAL were determined with a Michigan “O” (Hu-Friedy) periodontal probe using a flexible splint according to Isidor et al. (1984) so that measurements were made at the same site and in the same direction. Monthly measurements were made for one year. PPD and CAL were measured at 6 sites of each tooth: three buccal and three lingual or palatal. The data were collected when slight resistance to probe penetration was felt. When measurements were decimal, the number was rounded to the smaller whole number. Only the pockets measuring 4 mm or more were registered and evaluated monthly. The diabetic patients were submitted to monthly fasting glucose and glycosylated hemoglobin determinations for one year when the clinical measurements were made. At the end of the study, the diabetic group was divided into three subgroups, according to their metabolic state of diabetes, using glycosylated hemoglobin levels for reference (5.3-8.0%; Labtest Sistemas para Diagnostico, Belo Horizonte, Brazil): controlled patients: 5.3-8.0% (N = 9); moderately controlled patients: 8.1-9.0% (N = 13); poorly controlled patients: >9.0% (N = 8) (Glycohemoglobin test, Healthco International Inc., Boston, MA). Non-diabetic patients were submitted to glucose and glycosylated hemoglobin analyses during the first examination to verify the absence of diabetes. A fasting glucose determination was carried out during the last examination to confirm that the patient did not become diabetic during the study. The following statistical methodology recommended by Rodrigues (1986) and Campos (1983) was carried out: a) arithmetic means and standard deviations; b) linear correlation coefficient (r) and Student t-test; c) analysis of variance with Brieger’s F test was used to compare PPD and CAL, the glycosylated hemoglobin and the fasting glucose for the diabetic and control groups. When significance was shown, Tukey’s or Duncan’s test was used to determine significance at the 5% probability level.

Results

Tables 1 and 2 show the means ± SD of PPD, CAL, glycosylated hemoglobin and fasting glucose during the 12-month study in the diabetic and control groups, respectively.

Probing pocket depth (PPD)

The PPD of the diabetic group was 4.72 ± 0.16 mm and of the control group 4.65 ± 0.08 mm showing no significant difference (P>0.05) Analysis of variance was used to compare the means of the diabetic group in order to determine any statistically significant difference among the means for several months, since the value of the F test was significant (P<0.05). Considering this result, Tukey’s test showed significant differences (P<0.05) among the means of the 1st, 2nd, and 3rd months when compared to the 10th, 11th and 12th months of the study. Analysis of variance for PPD of the control group was not significant (P>0.05). Analysis of variance for PPD of the diabetic subgroups showed significant differences among the means (P<0.01) for the controlled patients during the 12 months. Tukey’s test showed significant differences (P<0.05) among the 1st and the 10th and 11th months. Analysis of variance showed significant differences (P<0.01) among the mean of the moderately controlled patients. Tukey’s test showed significant differences (P<0.05) when the mean of the 1st month was compared with the mean of the 4th month and so forth until the comparison of the mean for the 9th month with the mean for the 12th month. Analysis of variance for the 12 months for the poorly controlled patients showed significant differences (P<0.01) among the means. Tukey’s test showed significant differences (P<0.05) among the 1st, 2nd, 3rd, 4th, 5th, 6th, 7th and 8th months considering the means of other months.

Clinical attachment level (CAL)

The mean for CAL of the diabetic group was 8.34 ± 0.17 mm (Table 1), while the mean for CAL of the control group was 8.11 ± 0.11 mm with a statistically significant difference of 0.23 mm (P<0.01) (Table 2). Analysis of variance of the means of the diabetic group showed a significant result (P<0.05). Duncan’s test showed significant differences (P<0.05) among the mean for the 1st month when compared with the means for the 9th, 10th, 11th, and 12th months and among the mean of the 2nd month when compared with the 10th, 11th, and 12th months of the study. The means of the control group were not significant (analysis of variance, P>0.05). Analysis of variance for CAL of the diabetic subgroups showed significant differences among the means (P<0.01) for controlled patients during the 12 months. Tukey’s test showed significant differences (P<0.05) when the means for the 1st and 2nd months were compared with the 9th, 10th, and 11th months. Analysis of variance showed significant differences (P<0.01) among the means for the mildly controlled patients during the 12-month study. Tukey’s test showed significant differences (P<0.05) among the means for the 1st, 2nd, 3rd, and 4th months when compared with the 5th to 12th months of study, therefore, showing more differences among the means as compared with the previous group. Analysis of variance for the 12 months for the poorly controlled patients showed significant differences (P<0.01). Tukey’s test showed significant differences (P<0.05) among the means for the 1st to 7th months when compared with the means from the 4th to 12th months, showing therefore more differences among the means as compared with the former groups.

Glycosylated hemoglobin (GH)

GH levels of the diabetic group were 8.91 ± 2.08% (Table 1) and analysis of variance did not show significant differences (P>0.05). Analysis of variance comparing the means of GH among the diabetic subgroups (controlled patients, moderately controlled patients and poorly controlled patients) was signifi-cant (P<0.01); that is, the GH level increased from controlled to moderately controlled to poorly controlled patients. Tukey’s test showed that the com-parison among the means was significant (P<0.05).

Fasting glucose (FG)

FG level for the diabetic group was 151.79 ± 67.10 (Table 1) and analysis of variance did not show significant differences (P>0.05). The FG level increased significantly from the controlled to moderately controlled to the poorly controlled patients (analysis of variance, P<0.01). Tukey’s test showed that the comparison among the means was significant (P<0.05).

Comparison of diabetic and control groups

Table 3 shows the data resulting from the comparison between the diabetic and the control groups when the linear correlation coefficient (r) and the Student t-test were used. The t values at the 1% probability level were significant when the following correlations were made: a) PPD x CAL of the control group; b) PPD x CAL of the diabetic group. When the variables of the diabetic group were correlated, the results of the following comparisons showed significance (P<0.05): a) PPD x GH; b) CAL x GH; c) GH x FG. In the comparisons PPD x FG and CAL x FG, the t values were not significant. Analysis showed that for the control group and the subgroups of diabetic patients, the t values were significant (P<0.01) when correlating poorly controlled x control patients considering PPD and CAL (Table 4); controlled and moderately controlled patients when correlated with control patients were not significant (P>0.05).

Discussion

In this study no significant differences were found in the means of probing pocket depth for diabetic patients (NIDDM) and control patients (4.72 and 4.65 mm, respectively). During the year of study, a significant deepening (4.47 vs 4.94 mm) of the pockets was observed in the diabetic group. This was not observed in the control group which suggests a faster evolution of periodontal disease in diabetic patients. When comparing the diabetic group divided into subgroups with the control group, we noted that among the controlled and the moderately controlled diabetics and the control group, there were no significant differences. However, among the poorly controlled diabetics and the control group there were significant differences (P<0.01). Therefore, it is not sufficient to only evaluate if the patient is diabetic, but it is also necessary to evaluate the metabolic state of the disease. This could explain the fact that Benveniste et al. (1967) and Glavind et al. (1968) did not detect differences between the diabetic and control groups. Comparing the controlled, moderately controlled and poorly controlled subgroups, significant differences were observed in the probing pocket depths (P<0.05), emphasizing that it is important to determine the status of the diabetes. Correlating probing pocket depth and glycosylated hemoglobin in the NIDDM group, statistical significance was observed (P<0.05). However, comparison of probing pocket depth and fasting glucose was not statistically significant indicating that only the fasting glucose test may not be sufficient since it has been shown to be inadequate for this evaluation. This fact can also explain the differences of this study and those of Nichols et al. (1978) and Hove and Stallard (1970) which used fasting glucose. One means of measuring periodontal disease is the probing pocket depth, which in spite of the possible flaws, is still in use (Listgarten, 1980). Data concerning probing pocket depth of diabetic and non-diabetic patients showed that the differences were not significant (Benveniste et al., 1967; Glavind et al., 1968) and our results agree when considering the diabetic group as a whole, but our results disagree when the diabetic group is subdivided according to the metabolic status of the disease. Analyzing clinical attachment levels, Goodson et al. (1982) reported a dynamic condition for periodontal disease with periods of exacerbation and remission. Lindhe et al. (1983) and Harley et al. (1987) studied the progression of periodontal disease in adults in the absence of therapy and confirmed this result. In this study, the means of the clinical attachment levels of NIDDM patients were higher than those of the control patients (P<0.01). Means of the monthly clinical attachment levels of the diabetic group were significantly different (P<0.05), while those of the control group were not significantly different. This suggests a quicker evolution of periodontal disease in NIDDM patients. Subdividing the NIDDM group according to metabolic status of the disease, differences (P<0.01) were only observed between the poorly controlled diabetic group and the control group. When the monthly values of the subgroups were compared, significant differences were observed (P<0.05), showing that it is not sufficient to determine whether the patient is diabetic or not, but that it is very important to evaluate the metabolic status of the diabetes. Correlating the attachment level and glycosylated hemoglobin in the NIDDM group, statistical significance (P<0.05) was observed, while the correlation with fasting glucose was not significant. These results confirm the importance of determining the metabolic status of the diabetes and the importance of choice of test. Our results agree with numerous other studies: Cohen et al. (1970) observed greater attachment loss in diabetics than in non-diabetics; Sznajder et al. (1978), who studied attachment loss in diabetics and non-diabetics; showed greater loss in the diabetic group over 30 years of age, under similar circumstances; Shlossman et al. (1990), studying Pima Indians, reported that the average attachment loss was approximately three times greater in diabetic patients for all ages and both sexes; Emrich et al. (1991) reported that periodontal destruction measured by attachment loss and alveolar bone loss shown by radiography are greater and more prevalent in diabetic patients than in those with normal glucose tolerance. We would like to emphasize that literature on the analysis of the progression of periodontal disease in diabetes mellitus is scarce and that this study contributes to the evaluation of this progression during an entire year with monthly examinations and subdivision of diabetes mellitus to evaluate the influence of metabolic control of the disease. Several studies have found that the incidence and the severity of periodontal disease have been reported as greater in IDDM and NIDDM (Belting et al., 1964; Benveniste et al., 1967; Glavind et al., 1968; Cianciola et al., 1982; Novaes Jr. et al., 1991) and that rapidly progressive periodontal disease appears to be associated to poor metabolic control (Gislen et al., 1980). However, other investigators did not find any relationship between diabetes mellitus and periodontal disease (Hove and Stallard, 1970; Barnett et al., 1984). This incompatibility amongst studies can be partly due to differences in methodology that usually does not consider the metabolic status of the disease. Ervasti et al. (1985) did not find any differences in the periodontal status among diabetic groups as a whole and the control group. Thus, they subdivided the diabetic group into well controlled, moderately controlled and poorly controlled patients. Control was defined according to blood and urine glucose levels and glycosylated hemoglobin level. They reported a higher bleeding index in poorly controlled diabetics than in the other two subdivisions. According to our results, metabolic control plays an important role in analysis of the relationship between periodontal disease and diabetes mellitus. Metabolic control determined by glycosylated hemoglobin levels is more reliable than that determined by fasting glucose since this test is not dependent on cooperation of the patient, does not require fasting and is not affected by fluctuations of blood glucose on the day of examination. In conclusion, the systemic condition of diabetes mellitus (NIDDM) changes the evolution of periodontal disease. During the 12 months of study, alteration in the clinical attachment level was observed. Glucose level measured by fasting glucose did not show the same reliability as the glycosylated hemoglobin test.

References

Barnett ML, Yancy JM, MacMillan DR, Kotoyan M: Absence of periodontitis in a population of insulin dependent diabetes mellitus (IDDM) patients. J Periodont 55: 402-405, 1984

Belting CHM, Hiniker JJ, Dummett CO: Influence of diabetes mellitus on the severity of periodontal disease. J Periodont 35: 476-480, 1964

Benveniste R, Bixler D, Conneally PM: Periodontal disease in diabetics. J Periodont 38:5-13, 1967

Campos H: Estatística experimental não-paramétrica. 4th ed. ESALQ, USP, Piracicaba, 1983

Cianciola LJ, Park BH, Bruck E, Mosovich L, Genco RJ: Prevalence of periodontal disease in insulin-dependent diabetes mellitus (juvenile diabetes). J Am Dent Assoc 104: 653-660, 1982

Cohen DW, Friedman LA, Shapiro J, Kyle CC, Franklin S: Diabetes mellitus and periodontal disease: two years longitudinal observations Part 1. J Periodont 41: 709-712, 1970

Emrich LJ, Shlossman M, Genco RJ: Periodontal disease in non-insulin-dependent diabetes mellitus. J Periodont 62: 123-130, 1991

Ervasti T, Knuuttila M, Pohjamo L, Haukipuros K: Relation between control of diabetes and gingival bleeding. J Periodont 56: 154-157, 1985

Gislen G, Nilsson KO, Matsson L: Gingival inflammation in diabetic children related to degree of metabolic control. Acta Odont Scand 38: 241-246, 1980

Glavind L, Lund B, Loe H: The relationship between periodontal state and diabetes duration, insulin dosage and retinal changes. J Periodont 38: 341-347, 1968

Goodson JM, Tanner ACR, Haffajee AD, Sornberger GC, Socransky SS: Patterns of progression and regression of advanced destructive periodontal disease. J Clin Periodont 9: 472-481

Harley A, Floyd P, Watts T: Monitoring untreated periodontal disease. J Clin Periodont 14: 221-225

Hove KA, Stallard RE: Diabetes and the periodontal patient. J Periodont 41: 713-718, 1970

Isidor F, Karring T, Attstrom R: Reproductivity of pocket depth and attachment loss measurements when using a flexible splint. J Clin Periodont 11: 662-668, 1984

Lindhe J, Liljenberg B, Adielsson B: Effect of long-term tetracycline therapy on human periodontal disease. J Clin Periodont 10: 590-601, 1983

Listgarten MA: Periodontal probing: What does it mean? J Clin Periodont 7: 165-176, 1980

Nichols CH, Laster LL, Bodak-Gyovai LZ: Diabetes mellitus and periodontal disease. J Periodont 49: 85-88, 1978

Novaes Jr AB, Pereira ALA, de Moraes N, Novaes AB: Manifestations of insulin-dependent diabetes mellitus in the periodontium of young Brazilian patients. J Periodont 62: 116-122, 1991

Rhodus NL: Detection and management of the diabetic patient. Comp Cont Ed Dent 8: 73-78, 1987

Rodrigues PC: Bioestatística. 1st ed. Eduff, UFF, Niteroi, 1986

Shlossman M, Knowler WC, Pettit DJ, Genco RJ: Type 2 diabetes mellitus and periodontal disease. J Am Dent Assoc 121: 532-536, 1990

Sznajder N, Carraro JJ, Rugna S, Sereday M: Periodontal findings in diabetic and non-diabetic patients. J Periodont 49: 445-448, 1978

Correspondence: Arthur Belém Novaes Jr., Av. das Américas, 1155, Conj. 1002, 22631-000 Rio de Janeiro, RJ, Brazil.

Accepted March 6, 1996
Electronic publication: February, 1997

BACK TO CONTENTS