Relationship of the Two-Hour Blood Glucose Level to Two Oral Roentgenographic Findings

Introduction

A survey of the literature shows that many investigators tend to associate the roentgenographic findings of alveolar bone loss and periodontal widening with diabetes. There is an implication of a linear relationship between these roentgenographic findings and the diabetic state–the more pronounced the diabetes (blood glucose level) the more severe the alveolar bone loss and periodontal widening.

In a previous study¹ we analyzed the relation between oral roentgenographic findings and the glucose tolerance pattern. The evidence indicated that there is a linear relationship between the glucose tolerance pattern and the presence of alveolar bone loss and periodontal widening. In another study² we analyzed the relation between oral roentgenographic findings and fasting blood glucose levels. The results pointed to a parabolic relationship between the fasting blood glucose levels and the presence of alveolar bone loss and periodontal widening. Thus the total evidence suggested that both hyperglycemia and hypoglycemia are associated with alveolar bone loss and periodontal widening.

This report will analyze the relationship of the two roentgenographic signs (alveolar bone loss and periodontal widening) to the two-hour blood true-glucose level in a group of 100 routine dental patients. The two-hour concentration was selected because of the general belief that it is more representative of glucose metabolism than is the fasting blood glucose concentration.

Review of the Literature

Published material is available regarding the relationship between diabetes mellitus and: 1) alveolar bone loss, and 2) periodontal widening.

Alveolar bone loss. There are several reports on the association of this finding with the diabetic picture.^3-9 One investigator¹⁰ observed bone loss in 1 out of every 2 of a group of diabetic children. On the other hand, two teams^11,12 found no alveolar bone loss. Sheridan et al.¹³ observed normal alveolar bone in 24 per cent of patients with normal glucose tolerance patterns. They further observed that patients with either increased or decreased glucose tolerance showed a greater incidence of alveolar bone loss and that most of the diabetic patients exhibited considerable bone loss in contrast to other groups of patients having more physiologic glucose tolerance patterns.

Periodontal widening. Rutledge¹⁰ observed this widening in 14 of 20 juvenile diabetics. The findings of Rudy and Cohen¹² are in agreement with those of Rutledge. However, Ziskin et al.¹⁴ could not demonstrate periodontal widening in their series of cases. Sheridan et at.¹³ reported that, in patients with decreased glucose tolerance patterns (persons who would commonly be called diabetic), there was a 62 per cent incidence of periodontal widening. These investigators also observed that the incidence decreased with an increase in sugar tolerance.

Clinical Material and Methods

One hundred dental patients were studied in the Section on Oral Medicine at the University of Alabama School of Dentistry. As far as possible, the subjects were drawn at random. Table 1 shows the age distribution–almost a typical unimodal curve. There were 22 males and 78 females.

 

Table 1–Age Distribution

The roentgenograms on each patient were studied for possible alveolar bone loss and periodontal widening. Alveolar bone loss was rated on a 7-point scoring system in which zero was the score for a normal septal crest; a score of 1 was assigned for minimal alveolar bone loss; 2, for minimal to moderate alveolar bone loss; 3, when half the alveolar support was missing; 4, for moderate to severe alveolar bone loss; 5, for no alveolar support; and 6 for edentulous patients. Periodontal widening was scored on a 4-point scoring system in which zero was the score for no marginal widening; a score of 1 was given for minimal marginal widening; 2, for severe marginal widening; and 3 for edentulous patients.

A true-glucose tolerance test was performed on each patient according to the methods of Somogyi^15,16 and Nelson.¹⁷ No preparatory diet was recommended except for complete lasting for twelve hours prior to the laboratory examination. Complete details regarding the method of investigation are available in the publication by Sheridan et al.¹³ For this study, only the information derived from the two-hour blood glucose determination was utilized.

Results

The two-hour blood glucose levels were considered in relation to: 1) each roentgenographic sign, singly, and 2) a combination of the two signs.

Single Signs

Alveolar bone loss: This roentgenographic sign was assessed in terms of: 1) mean scores, 2) coefficients of variation, and 3) statistical significance.

The scores for edentulous patients were included in this analysis since this condition would be an end-result of alveolar bone loss. Although loss of teeth can be due to other causes, the exclusion of these cases would be too discriminatory against those in which tooth loss was due to alveolar bone loss. The available data do not permit the selection of cases in which loss of teeth was accompanied by severe alveolar bone loss.

Mean scores. Examination of Figure 1 shows that alveolar bone loss was least in the patients with a two-hour blood glucose level of 80-89 mg. per 100 ml. The bone loss scores increased at the blood glucose levels of <70, 70-79, and 90-99 mg. per 100 ml., and there was a marked increase at the level of >100 mg. per 100 ml. Figure 1 also shows a comparison with the mean scores obtained on dentulous cases alone. The plotted mean scores for these cases graphically illustrate how these values can be affected by the exclusion of patients who have lost all of their teeth. It is noteworthy that of the 11 edentulous patients in this study not one had a two-hour blood glucose level in the range of 80-89 mg. per 100 ml.

Figure 1

 

Coefficients of variation. Examination of Table 2 reveals that the lowest coefficients of variation were observed at blood glucose levels of 80-89 and >100 mg. per 100 ml. Thus, in these groups the individual values seemed to cluster about the mean more than in the other blood glucose groups.

 

Table 2–Relationship of Two-Hour Blood Glucose Level to Alveolar Bone Loss

Statistical significance. From Table 2 it can be seen that there was no statistically significant difference between any of the blood glucose groups compared to similar groups with levels higher or lower by 10 mg. per 100 ml. This is shown by the probability values of .5 and >.2. The P value was also checked for all the groups. There were no statistically significant differences except when the 80-89 and >100 mg. per 100 ml. groups were compared. This is shown by a P value of <.005.

Periodontal widening: This roentgenographic sign was also assessed in terms of: 1) mean scores, 2) coefficients of variation, and 3) statistical significance.

Mean scores. Figure 2 shows the mean scores for all patients (dentulous and edentulous) and the mean scores when edentulous patients were excluded. Examination of these values shows the effect of excluding the edentulous cases. Patients with severe periodontal widening leading to edentulousness would not be counted if only dentulous cases were considered. This effect was greatest in the <70 and the 90-99 mg. per 100 ml. blood glucose groups. The rest of this analysis pertains to the scores obtained from the study of all the patients. Figure 2 shows that periodontal widening was least in the group with blood glucose levels of 80-89 mg. per 100 ml., and that the incidence of this condition increased as the two-hour blood glucose level rose above or fell below this small range.

Figure 2

 

Coefficients of variation. From Table 3 it may be seen that the lowest coefficients of variation were shown by the groups with blood glucose levels of 80-89 and >100 mg. per 100 ml. That is, the individual scores were clustered more closely about the mean in these two groups.

 

Table 3–Relationship of Two-Hour Blood Glucose Level to Periodontal Widening

Statistical significance. From Table 3 it is evident that there were no statistically significant differences between any of the blood glucose groups when compared to similar groups with levels higher or lower by 10 mg. per 100 ml. This is shown by probability values that are greater than .05. The P value was also checked for all the groups. There were no statistically significant differences shown except when the 80-89 and >100 groups were compared. A P value of <.05 so derived is statistically significant.

Combination of the Two Signs

An analysis of the data was made for alveolar bone loss and periodontal widening. The combination of these two signs was assessed in terms of: 1) mean scores, 2) coefficients of variation, and 3) statistical significance. There was a difference in the scoring systems for these two signs, alveolar bone loss being measured on a 7-point scale, and periodontal widening on a 4-point scale. In order to equate these scores, the periodontal scores were doubled, so that a value could be placed on the combination of these two roentgenographic signs.

Mean scores. It appears from Figure 3 that the relationship between the two-hour blood glucose level and the presence of these two signs is parabolic. Thus alveolar bone loss and periodontal widening were least when the two-hour blood levels were in the range of 80-89 mg. per 100 ml. As the two-hour blood glucose level increased or decreased, the frequency of alveolar bone loss combined with periodontal widening increased.

Figure 3

 

CoeffIcients of variation. Table 4 shows that the lowest coefficients of variation occurred in the groups with blood glucose levels of 80-89 and >100 mg. per 100 ml. This indicates that the individual scores clustered more closely around the mean in these two blood glucose groups.

 

Table 4–Relationship of Two-Hour Blood Glucose Level to Both Alveolar Bone Loss and Periodontal Widening

Statistical significance. As seen in Table 4, there were no statistically significant differences between any of the two-hour blood glucose groups and their adjacent groups. A comparison of each group with all others showed no statistically significant differences except between the >100 mg. group and the 70-79 and 80-89 mg. groups. The P values here were <.05 and <.01 respectively.

Discussion

The data will be discussed under the following headings: 1) mean scores, 2) coefficients of variation, and 3) statistical probability values. Because of the close correlation between age and glucose levels, and between age and roentgenographic signs, the age factor will be analyzed in terms of blood glucose concentration and roentgenographic signs.

Mean scores

An analysis of the mean scores for the 2 oral roentgenographic signs, both independently and in combination, point up the fact that the patterns are parabolic. The data indicate that the range of blood glucose normality associated with the lowest incidence of one or both of these particular roentgenographic signs (Table 5) was in the general range of 80-89 mg. per 100 ml. Moreover, two-hour blood glucose levels above and below this narrow range were apt to be associated with a high incidence of abnormal roentgenographic findings.

 

Table 5–Comparison of Mean Scores

Coefficients of variation

Generally speaking, the lower the coefficient of variation, the more the individual scores cluster about the mean. Thus, a coefficient of variation of zero is the ultimate, in that it signifies that all subjects provide scores exactly like the mean score.

In Table 6 are summarized the coefficients of variation with respect to the two-hour blood glucose groups and the single and combined oral roentgenographic signs. It can be seen that the individual scores always clustered more closely about the mean in the groups with blood glucose levels of 80-89 and >100 mg. per 100 ml.

 

Table 6–Comparison of Coefficients of Variation

Significant probability factors

Table 7 contains a tabulation of the significant probability findings for the single and combined oral roentgenographic signs. It is noteworthy that patients with two-hour, blood glucose levels of >100 mg. per 100 ml. showed significant differences in the amount of alveolar bone loss and periodontal widening when compared to patients with glucose levels of 80-89 mg. per 100 ml. The significant difference between these blood glucose groups was maintained when the data for both roentgenographic signs were combined. Moreover, a significant difference was also shown between the 70-79 and >100 mg. groups for the combination of alveolar bone loss and periodontal widening. It is of interest that the two-hour blood glucose groups (80-89 and >100 mg. per 100 ml.) that consistently showed significant differences also demonstrated the lowest coefficients of variation.

 

Table 7–Comparison of Significant Probability Values

Age factor

Some investigators contend that, with increasing age the values for normal blood glucose concentration increase. Certainly, the data from which this study is derived indicate that oral roentgenographic signs occur predominantly in the older age groups (Table 8). Hence, it was thought necessary to analyze the relationship between roentgenographic signs and blood glucose levels in a single age group.

 

Table 8–Percentage Frequency of Patients with Oral Roentgenographic Signs, by Age Groups

Figure 4 pictorially demonstrates the relationship of two-hour blood glucose concentration to alveolar bone loss and periodontal widening in the 19 patients between the ages of 40 and 49. Examination of Figure 4 and a comparison of these data with those in Figure 3 indicate that the parabola is more exaggerated when the age factor is eliminated. A comparison of Tables 4 and 9 shows that the individual scores for this one age group clustered more about the mean than did the scores for the entire group as judged by the coefficients of variation. There were no significant differences when each group was compared to any other, except between the 80-89 mg. and >100 mg. blood glucose groups; here the P value was <.05. It is noteworthy that, although these blood glucose groups still showed a significant difference, there was some loss in the P value (from <.0 5 to <.01). It thus appears that age or some factor operating with age may be involved in the relationship of the two oral roentgenographic signs and glucose metabolism. However, this factor is not the only, or even the dominant one.

Figure 4

 

Table 9–Relationship of Two-Hour Blood Glucose Level to Alveolar Bone Loss and Periodontal Widening in the 40-49 Year Age Group

Summary

An analysis is presented of the relationship between two oral roentgenographic signs (alveolar bone loss and periodontal widening) and the two-hour blood true-glucose level in 100 routine dental patients.

Comparison of the mean score values for the two roentgenographic signs, independently and when combined, demonstrated that the patterns are generally parabolic.

The coefficients of variation showed that, in general, the individual scores clustered more closely about the mean in the groups with blood glucose levels of 80-89 and >100 mg. per 100 ml.

The significant probability values demonstrated a significant difference between the 80-89 and >100 mg. two-hour blood glucose groups for both of the oral roentgenographic signs, singly and in combination.

The findings suggest that the two roentgenographic signs (alveolar bone loss and periodontal widening) are associated with both hyperglycemia and hypoglycemia.

An age factor or some factor operating with age seems to be associated with hyperglycemia and hypoglycemia in the relationship between the two roentgenographic signs and glucose metabolism. However, this factor does not play the only, or the dominant role in the relationship.

It is concluded that the range of normality for the two-hour blood glucose level with respect to these two roentgenographic signs is more narrow than usually accepted.

The fasting blood glucose level (previously reported), rather than the two-hour level, is more representative of glucose metabolism in relation to alveolar bone loss and periodontal widening.

 

References Cited:

1. Cheraskin, E., and Manson-Hing, L.: “The normal glucose tolerance pattern: development of blood glucose normality by an analysis of oral roentgenographic findings,” Oral Surg. 13: 819-835 (July) 1960.
2. Manson-Hing, L., and Cheraskin, E.: “The relationship of fasting blood glucose to oral roentgenographic findings,” J. Roentgenol. (in press).
3. Rudy, A., and Cohen, M. M.: “The oral aspects of diabetes mellitus,” New England J. Med. 219: 503-508 (Oct. 6) 1938.
4. Kaplan, N.: “Oral symptomatology of diabetes mellitus,” Pennsylvania Dent. J. 41: 50- 51; 60-61 (Nov.) 1938.
5. Pollack, H.; Person, P., and Knishkowy, E.: “Diabetes mellitus, its relation to oral pathosis. II. Oral lesions seen in diabetes mellitus,” Periodont. 18: 155-158 (Oct.) 1947.
6. Martinez, E.: “Diabetes in dental practice,” Abstr. 1: 178 (March) 1956.
7. Stahl, S. S.: “Roentgenologic and bacteriologic aspects of periodontal changes in diabetes,” Periodont. 19: 130-132 (Oct.) 1948.
8. Golomb, I. M.: “An evaluation of the relation of diabetes mellitus to periodontal disease,” New York State Dent. J.15: 525-528 (Nov.) 1959.
9. Person, P.: “Some aspects of bone metabolism of the maxilla and mandible,” New York State Dent. J. 23: 165-170 (Apr.) 1957.
10. Rutledge, C. E.: “Oral and roentgenographic aspects of the teeth and jaws of juvenile diabetics,” A. D. A. 27: 1740-1750 (Nov.) 1940.
11. Lovestedt, S. A., and Austin, L. T.: “Periodontoclasia in diabetes mellitus,” A. D. A. 30: 273-275 (Feb.) 1943.
12. Rudy, A., and Cohen, M. M.: “Oral aspects of diabetes mellitus,” A. D. A. 29: 523-534 (Apr.) 1942.
13. Sheridan, R. C., Jr.; Cheraskin, E.; Flynn, F. H., and Hutto, A. C.: “Epidemiology of diabetes mellitus. II. A study of 100 dental patients,” Periodont. 30: 298-323 (Oct.) 1959.
14. Ziskin, D. E.; Siegel, E. H., and Loughlin, W. C.: “Diabetes in relation to certain oral and systemic problems. Part I. Clinical study of caries, tooth eruption, gingival changes, growth phenomena and related observations in juveniles,” Dent. 23: 317- 331 (Oct.) 1944.
15. Somogyi, M.: “A new reagent for the determination of sugars,” Biol. Chem. 160: 61-68 (Sept.) 1945.
16. Somogyi, M.: “Determination of blood sugar,” Biol. Chem. 160: 69-73 (Sept.) 1945.
17. Nelson, N.: “A photometric adaptation of the Somogyi method for the determination of blood sugar,” Biol. Chem. 163: 375-380 (May) 1944.