Comminution of Food by Complete-denture Wearers A.P. SLAGTER, L.W. OLTHOFF, W.H.A. STEEN, and F. BOSMAN Department of Oral Maxillofacial Surgery, Prosthodontics and Special Dental Care, Faculty of Medicine, University of Utrecht, Padualaan 14, 3584 CH Utrecht, The Netherlands
A method for quantifying the comminution of an artificial test food (Optosil®) was evaluated with respect to its suitability for measurements of masticatory performance in complete-denture wearers. Reference was made to subjects with natural dentitions. The description of particle size distributions generated by complete-denture wearers by a Rosin-Rammler equation was subject to limitations, due to the presence of large proportions of almost-intact particles, which had hardly been damaged or broken during chewing. This finding might be explained by: (i) the relatively high fracture strength of Optosil® as compared with natural foods; and (ii) the limitations of complete-denture wearers in exerting bite forces. The particle size distributions obtained after repeated measurements and described by linear interpolation of data points were reproducible. In comparison with young adults with natural dentitions, the denture-wearers needed approximately seven times more chewing strokes to achieve an equivalent reduction in particle size.
Olthoffet al. (1984), but now modified with respect to the size ofthe Optosil® cubes and the amount of test food offered, and (ii) to compare the comminution of Optosil' by complete-denture wearers with that by young adults with natural dentitions.
Materials and methods.
Subjects.-Thirteen persons wearing complete dentures (mean age, 61 years; range, 48-69 years) and five persons with complete natural dentitions (mean age, 26 years; range, 23-29 years) participated in the experiments, after giving their informed consent. Experimental procedure.-Masticatory performance was evaluated in a series of chewing tests described in detail by Olthoff et al. (1984). In the present study, however, both the sizes ofthe particles and the portion of test food were reduced. All subjects were offered cubes of Optosil@ with an edge size of 5.6 mm in portions of 17 particles (approximately 3 cm3). The test food was collected after 20, 40, 60, and 80 chewing strokes. The procedure was carried out twice J Dent Res 71(2):380-386, February, 1992 for each number of strokes, and the two portions of chewed test food were pooled. The particles were sieved on a stack of up to 10 sieves, Introduction. with apertures from 5.6 decreasing to 0.5 mm, with a bottom plate. The amount of test food on each sieve and the bottom plate was Masticatory performance is frequently evaluated by means of tests, weighed. based upon the chewing of a comminutable test food. Most tests Measurements.-The chewing tests were carried out twice with described in the literature use natural foods and one or more sieves intervals of one week (Table 1). to measure the degree of comminution (e.g., Manly and Braley, Data analysis.-Each particle size distribution obtained after 1950; Manly and Vinton, 1951; Yurkstas, 1954; Kapur and Soman, completion of a specific number of chewing strokes was described by 1964; Helkimo et al., 1978; Jiffry, 1981). plotting the percentage of the test food by weight that had passed a The physical properties of a natural food, such as force-deforma- certain sieve, the cumulative percentage undersize [Q, (X)], as a tion characteristics and fracture strength, as well as the shapes and function of the logarithm of the sieve aperture (X). A non-linear sizes of its particles may vary. The use of an artificial test food may regression procedure (Bevington, 1969) was applied to obtain indireduce these sources of variation. A silicone rubber (Optosil®, Bayer vidual estimates for the two parameters (X50 and b) in the RosinAG, Leverkusen, Germany; version 1980) has been chosen because Rammler equation: its form and physical properties, as well as the results of comminution, are highly reproducible (Edlund and Lamm, 1980; Olthoff et (1) Q, (X) = 100 [1 - 2(x1x5&b] al., 1984, 1986). Many investigators have chosen only one (Manly and Braley, The median particle size (X50) is the aperture of a theoretical sieve 1950; Manly and Vinton, 1951; Yurkstas, 1954; Kapur and Soman, through which 50% ofthe test food particles by weight can pass. The 1964), some two (Edlund and Lamm, 1980) or three sieves (Helkimo parameter b indicates the extent to which the particles are equally et al., 1978) and a more or less arbitrary index to characterize the sized. The individual estimates of X50 can be averaged for the result ofthe chewing process. However, when more sieves are used, complete-denture wearers and for the dentate subjects, giving the more detailed information is gained on the distribution of particle mean values of X50 for each group. In another procedure, the sizes in chewed food. Some authors have used up to ten sieves cumulative percentage undersize values from all subjects in a group (Lucas and Luke, 1983; Olthoffet al., 1984; Lucas et al., 1986; Van were pooled for each sieve aperture. The non-linear regression der Bilt et al., 1987) or more (Jiffry, 1981, 1983). Olthoffet al. (1984) procedure was carried out on these pooled data so that group found that the particle size distributions generated by dentate estimates of X50 and b could be obtained. subjects could be described with two parameters by a cumulative Since the data points of particle size distributions from indidistribution function, according to Rosin and Rammler (1933). vidual complete-denture wearers were not described well by the In a preliminary study of masticatory performance in denture- Rosin-Rammler equation, another approach was chosen for characwearers, replication of the test described by Olthoff et al. (1984) terization of the size distributions (Lucas et al., 1986). The theoretiindicated that cubic Optosil particles with an 8-mm edge size, cal sieve apertures through which 20% (X20), 50% (X50), and 80% offered in portions of 4 cm3, were very difficult to chew. The particles of the particles by weight would have passed were estimated(X80) by remained mostly intact, and the Rosin-Rammler equation failed to linear interpolation. The ratio to indicate the was used X2AX describe the size distributions. degree of variation of particle sizes within the distribution. The aims of this study were (i) to determine the suitability for The individual values of X50 were plotted vs. the number of denture-wearers of the masticatory performance test described by chewing strokes (N), on a double logarithmic scale. The relationship between X50 and N could be described by the equation: Received for publication January 15, 1991 Accepted for publication October 1, 1991
380
X50(N) = (Xo cC Nd)/(Xo + c Nd)
Downloaded from jdr.sagepub.com at UNIV OF PITTSBURGH on December 18, 2014 For personal use only. No other uses without permission.
(2)
FOOD COMMINUTION WITH COMPLETE DENTURES
Vol. 71 No. 2
where c and d are two variables describing the rate of reduction of X50 as a function of the number of chewing strokes, and XO is a constant, depending on the initial particle size. The values of c and d were estimated for each subject by a non-linear regression procedure (Bevington, 1969) during which X0 was assigned a fixed value of 5.6 mm. Then, the values of c and d being known, the number of chewing strokes necessary to reduce the value of X50 to halfthe initial particle size (2.8 mm) was calculated: N 1/2 In addition, group estimates of c and d were obtained by pooling the data from all subjects in each group. With these group estimates, the values of N ,2 were calculated for the dentate and the complete-denture groups. An overview ofthe methods and variables describing the particle size distributions and the reduction in median particle size during the chewing process is given in Table 2. Statistical analysis.-The procedure for assessing differences in particle size distributions between the complete-denture and the
381
dentate groups, within individuals, and between individuals within both groups started with the Kolmogorov-Smirnov one-sample test for testing the distributions of the variables for normality (Massey, 1951). The variables were also tested for equality ofinter-individual variances by comparison of the complete-denture with the dentate group by means of the F-ratio test. Differences between completedenture wearers and dentate subjects were tested by Student's t test for independent samples in the case of normal distributions and equal inter-individual variances, by Student's t test with a correction of the degrees of freedom in the case of unequal inter-individual variances (Welch, 1947), and by the Mann-Whitney U-test in the case of non-normality. Differences in paired data within each group were tested by Student's t test for paired samples with normal distributions and by Wilcoxon's matched-pairs signed-ranks test in the case of non-normality. Table 3 lists the formulae for calculation of the inter-individual and intra-individual variances of the variables.
TABLE 1 THE MASTICATORY PERFORMANCE TESTS SESSIONS, NUMBERS OF CHEWING STROKES, NUMBERS OF PARTICIPANTS Numbers of participants Numbers of chewing strokes
Session I
Session II
20
10
10
40
13
13
60
13
13
80
13
13
20
0
5
40
5
5
60
5
5
80
5
5
Complete-denture group
Dentate group
TABLE 2 THE METHODS AND VARIABLES DESCRIBING THE PARTICLE SIZE DISTRIBUTIONS AND THE REDUCTION IN PARTICLE SIZE DURING CHEWING
Method
Symbol
Variable
Weighing
Qw (X)
The cumulative percentage undersize: the percentage of test food by weight that can pass a certain sieve aperture. Sieve aperture or particle size.
X
Curve-fitting equation (1)
X50 b
Linear interpolation
X20 X50 X80
Observation and counting
X20/X80 N
Curve-fitting equation (2)
X0 c, d
Calculation from equation (2)
N112
The aperture of a theoretical sieve for which QV = 50%, the median particle size. A measure of the extent to which the particles are equally sized. The apertures of theoretical sieves for which Q, = 20%, 50%, and 80%, respectively. A measure of the extent to which the particles are equally sized. The number of chewing strokes. The initial particle size Measures of the rate of reduction in median particle size. The number of chewing strokes for which X50 = 0.5 X0.
Equation (1): Q, (X) = 100 [1 - 2-xx-)b] Equation (2): X50(N) = (X0 c
Nd)/(Xo + c Nd)
Downloaded from jdr.sagepub.com at UNIV OF PITTSBURGH on December 18, 2014 For personal use only. No other uses without permission.
SLAGTER et al.
382
J Dent Res February 1992
TABLE 3 THE STATISTICAL FORMULAE USED FOR DETERMINATION OF THE INTRA- AND INTER-INDIVIDUAL VARIABILITY OF X20, X50, X80, c, d, AND N1/2
Formula
Purpose to estimate the intra-individual variance in the particle size distributions obtained from two sessions in terms of X20, X50, and X80 for each subject to estimate the intra-individual variance in the reduction in median particle size in terms of c, d, and N112 for each subject to calculate the relative intra-individual variation in each of the variables for each subject, expressed as a percentage
si, intra = (xi'l xi,2)) / 2
Sintra / [(x11 + x,) / 2] 100
X/2 +/ il i,2)
inter i== (( s2
x
(x
+x
)/2
)to estimate the inter-individual variance in each of the variables
n
for both the complete-denture wearers and the dentate subjects
n-1
Results. Fig. 1 shows the cumulative size distributions obtained from the pooled data of the denture-wearers and the dentate subjects after 20 chewing strokes. Estimates of X50 and b were not obtained, since the Rosin-Rammler equation did not adequately describe the data points. In Fig. 2, the curves representbestfits ofthe Rosin-Rammler equation for the pooled data. In Figs. 1 and 2, the mean, minimum, and maximum values for the individual estimates ofX20, X50, and X.0 are indicated as well. The group estimates of X50 and b and their corresponding 95% confidence intervals are given in Table 4. The dentate subjects and the denture-wearers differed markedly with respect to X20, X5 and X80, but overlapped with respect to the parameter b. All dentate subjects were able to comminute the test food particles to such an extent that after 20 strokes no particles were left on the upper sieve, with an aperture of 5.6 mm. Most denturewearers did not achieve a reduction in particle size to that extent, 100 -
Qw (%)
and even after 80 chewing strokes, up to 40% of the test food material by weight could be found on that sieve. In Table 5, the mean, median, minimum, and maximum percentages of the particles by weight that remained on the 5.6-mm sieve are given for each number of chewing strokes. The values for the parameters (X20, X50, X80, X20/X80, c, d, and did not differ significantly from a normal distribution N1/2) (Kolmogorov-Smirnov test). Differences in values of X20, X50, and X50 between denturewearers and dentate subjects were extremely significant (p < 0.0001). The mean values and standard deviations for both groups are given in Table 6. The inter-individual variance in the group of denturewearers was significantly larger than that in the group of dentate subjects (p < 0.05). One-tailed probability values of the differences in X53, X50, and X80 between 20,40, 60, and 80 chewing strokes were more significant in the group of denture-wearers (p < 0.005) than in the dentate group (p < 0.02). Between 20 and 80 chewing strokes, the reduction in X50 100 1
Qw (%) _
80 1
80 -
XI
L
0// 60 -
60 -
4X
LI
40
+
40
n
20
Li
0 10
0.1
x (mm)
Fig. 1-The cumulative percentage undersize by weight [Q, (X)] plotted
as a function of the sieve aperture (X) after 20 chewing strokes. The mean
values of X20, X50, and X80 for the complete-denture and the dentate groups are represented by an *. The minimum and maximum values of X20, X50, and X are indicated by a Dl for the complete-denture and a O for the dentate group. The curves represent the cumulative size distributions of each group: The dotted curve indicates the complete-denture and the dashed curve the dentate group.
Li
0 /,
20
-.0.1
.
( 1
10
x (mm)
Fig. 2-The cumulative percentage undersize by weight [Q, (X)] plotted
as a function of the sieve aperture (X) after 80 chewing strokes. The mean values of X20, X50, and X80 for the complete-denture and the dentate groups are represented by an *. The minimum and maximum values ofX20, X50, and are indicated by a En for the complete-denture and a O for the dentate Y.. group. The curves were obtained by curve-fitting the pooled data of each group, according to the Rosin-Rammler equation (1). The dotted curve indicates the complete-denture and the dashed curve the dentate group.
Downloaded from jdr.sagepub.com at UNIV OF PITTSBURGH on December 18, 2014 For personal use only. No other uses without permission.
383
FOOD COMMINUTION WITH COMPLETE DENTURES
Vol. 71 No. 2
TABLE 4 THE GROUP ESTIMATES AND 95% CONFIDENCE INTERVALS OF X50, b, c, AND d OBTAINED FROM THE POOLED DATA OF THE COMPLETE-DENTURE WEARERS AND THE DENTATE SUBJECTS AND THE CALCULATED VALUES OF N12 Number of chewing strokes
Complete-denture group
X50 (mm)
b
20 40
3.88
(3.67 - 4.09)
2.34
(1.89 - 2.79)
60
3.23
(3.03 - 3.44)
2.02
(1.65 - 2.39)
80
2.64
(2.47 - 2.82)
1.95
(1.62 - 2.28)
c (mm)
d
3983 (-3104 - 11,070)
-1.50 (-1.94 - -1.07)
X. (mm)
b
40
1.35 (1.29 - 1.42)
1.80
(1.59 - 2.01)
60
1.07 (1.04 - 1.09)
1.91
(1.81 - 2.01)
80
0.95 (0.93 - 0.97)
2.00
(1.89 - 2.11)
c (mm)
d
53.3 (34.9 - 71.7)
-0.91 (-1.00 - -0.81)
N1,2= 80 Number of chewing strokes Dentate group
20
N12 = 12 larger in the complete-denture wearers. The ratio X./X8 decreased significantly from 20 to 60 chewing strokes in the denture-wearers (p < 0.02), indicating that the particles were distributed over more different size classes as the number of chewing strokes increased. The ratio X,2X80 showed no significant change as a function ofthe number of chewing strokes in the dentate group. The intra-individual differences between sessions inX20, X50, and X80 were not significant for any number of chewing strokes (0.08 < p
A method for quantifying the comminution of an artificial test food (Optosil®) was evaluated with respect to its suitability for measurements of masticatory performance in complete-denture wearers. Reference was made to subjects with natural dentitions. The description of particle size distributions generated by complete-denture wearers by a Rosin-Rammler equation was subject to limitations, due to the presence of large proportions of almost-intact particles, which had hardly been damaged or broken during chewing. This finding might be explained by: (i) the relatively high fracture strength of Optosil® as compared with natural foods; and (ii) the limitations of complete-denture wearers in exerting bite forces. The particle size distributions obtained after repeated measurements and described by linear interpolation of data points were reproducible. In comparison with young adults with natural dentitions, the denture-wearers needed approximately seven times more chewing strokes to achieve an equivalent reduction in particle size.
Olthoffet al. (1984), but now modified with respect to the size ofthe Optosil® cubes and the amount of test food offered, and (ii) to compare the comminution of Optosil' by complete-denture wearers with that by young adults with natural dentitions.
Materials and methods.
Subjects.-Thirteen persons wearing complete dentures (mean age, 61 years; range, 48-69 years) and five persons with complete natural dentitions (mean age, 26 years; range, 23-29 years) participated in the experiments, after giving their informed consent. Experimental procedure.-Masticatory performance was evaluated in a series of chewing tests described in detail by Olthoff et al. (1984). In the present study, however, both the sizes ofthe particles and the portion of test food were reduced. All subjects were offered cubes of Optosil@ with an edge size of 5.6 mm in portions of 17 particles (approximately 3 cm3). The test food was collected after 20, 40, 60, and 80 chewing strokes. The procedure was carried out twice J Dent Res 71(2):380-386, February, 1992 for each number of strokes, and the two portions of chewed test food were pooled. The particles were sieved on a stack of up to 10 sieves, Introduction. with apertures from 5.6 decreasing to 0.5 mm, with a bottom plate. The amount of test food on each sieve and the bottom plate was Masticatory performance is frequently evaluated by means of tests, weighed. based upon the chewing of a comminutable test food. Most tests Measurements.-The chewing tests were carried out twice with described in the literature use natural foods and one or more sieves intervals of one week (Table 1). to measure the degree of comminution (e.g., Manly and Braley, Data analysis.-Each particle size distribution obtained after 1950; Manly and Vinton, 1951; Yurkstas, 1954; Kapur and Soman, completion of a specific number of chewing strokes was described by 1964; Helkimo et al., 1978; Jiffry, 1981). plotting the percentage of the test food by weight that had passed a The physical properties of a natural food, such as force-deforma- certain sieve, the cumulative percentage undersize [Q, (X)], as a tion characteristics and fracture strength, as well as the shapes and function of the logarithm of the sieve aperture (X). A non-linear sizes of its particles may vary. The use of an artificial test food may regression procedure (Bevington, 1969) was applied to obtain indireduce these sources of variation. A silicone rubber (Optosil®, Bayer vidual estimates for the two parameters (X50 and b) in the RosinAG, Leverkusen, Germany; version 1980) has been chosen because Rammler equation: its form and physical properties, as well as the results of comminution, are highly reproducible (Edlund and Lamm, 1980; Olthoff et (1) Q, (X) = 100 [1 - 2(x1x5&b] al., 1984, 1986). Many investigators have chosen only one (Manly and Braley, The median particle size (X50) is the aperture of a theoretical sieve 1950; Manly and Vinton, 1951; Yurkstas, 1954; Kapur and Soman, through which 50% ofthe test food particles by weight can pass. The 1964), some two (Edlund and Lamm, 1980) or three sieves (Helkimo parameter b indicates the extent to which the particles are equally et al., 1978) and a more or less arbitrary index to characterize the sized. The individual estimates of X50 can be averaged for the result ofthe chewing process. However, when more sieves are used, complete-denture wearers and for the dentate subjects, giving the more detailed information is gained on the distribution of particle mean values of X50 for each group. In another procedure, the sizes in chewed food. Some authors have used up to ten sieves cumulative percentage undersize values from all subjects in a group (Lucas and Luke, 1983; Olthoffet al., 1984; Lucas et al., 1986; Van were pooled for each sieve aperture. The non-linear regression der Bilt et al., 1987) or more (Jiffry, 1981, 1983). Olthoffet al. (1984) procedure was carried out on these pooled data so that group found that the particle size distributions generated by dentate estimates of X50 and b could be obtained. subjects could be described with two parameters by a cumulative Since the data points of particle size distributions from indidistribution function, according to Rosin and Rammler (1933). vidual complete-denture wearers were not described well by the In a preliminary study of masticatory performance in denture- Rosin-Rammler equation, another approach was chosen for characwearers, replication of the test described by Olthoff et al. (1984) terization of the size distributions (Lucas et al., 1986). The theoretiindicated that cubic Optosil particles with an 8-mm edge size, cal sieve apertures through which 20% (X20), 50% (X50), and 80% offered in portions of 4 cm3, were very difficult to chew. The particles of the particles by weight would have passed were estimated(X80) by remained mostly intact, and the Rosin-Rammler equation failed to linear interpolation. The ratio to indicate the was used X2AX describe the size distributions. degree of variation of particle sizes within the distribution. The aims of this study were (i) to determine the suitability for The individual values of X50 were plotted vs. the number of denture-wearers of the masticatory performance test described by chewing strokes (N), on a double logarithmic scale. The relationship between X50 and N could be described by the equation: Received for publication January 15, 1991 Accepted for publication October 1, 1991
380
X50(N) = (Xo cC Nd)/(Xo + c Nd)
Downloaded from jdr.sagepub.com at UNIV OF PITTSBURGH on December 18, 2014 For personal use only. No other uses without permission.
(2)
FOOD COMMINUTION WITH COMPLETE DENTURES
Vol. 71 No. 2
where c and d are two variables describing the rate of reduction of X50 as a function of the number of chewing strokes, and XO is a constant, depending on the initial particle size. The values of c and d were estimated for each subject by a non-linear regression procedure (Bevington, 1969) during which X0 was assigned a fixed value of 5.6 mm. Then, the values of c and d being known, the number of chewing strokes necessary to reduce the value of X50 to halfthe initial particle size (2.8 mm) was calculated: N 1/2 In addition, group estimates of c and d were obtained by pooling the data from all subjects in each group. With these group estimates, the values of N ,2 were calculated for the dentate and the complete-denture groups. An overview ofthe methods and variables describing the particle size distributions and the reduction in median particle size during the chewing process is given in Table 2. Statistical analysis.-The procedure for assessing differences in particle size distributions between the complete-denture and the
381
dentate groups, within individuals, and between individuals within both groups started with the Kolmogorov-Smirnov one-sample test for testing the distributions of the variables for normality (Massey, 1951). The variables were also tested for equality ofinter-individual variances by comparison of the complete-denture with the dentate group by means of the F-ratio test. Differences between completedenture wearers and dentate subjects were tested by Student's t test for independent samples in the case of normal distributions and equal inter-individual variances, by Student's t test with a correction of the degrees of freedom in the case of unequal inter-individual variances (Welch, 1947), and by the Mann-Whitney U-test in the case of non-normality. Differences in paired data within each group were tested by Student's t test for paired samples with normal distributions and by Wilcoxon's matched-pairs signed-ranks test in the case of non-normality. Table 3 lists the formulae for calculation of the inter-individual and intra-individual variances of the variables.
TABLE 1 THE MASTICATORY PERFORMANCE TESTS SESSIONS, NUMBERS OF CHEWING STROKES, NUMBERS OF PARTICIPANTS Numbers of participants Numbers of chewing strokes
Session I
Session II
20
10
10
40
13
13
60
13
13
80
13
13
20
0
5
40
5
5
60
5
5
80
5
5
Complete-denture group
Dentate group
TABLE 2 THE METHODS AND VARIABLES DESCRIBING THE PARTICLE SIZE DISTRIBUTIONS AND THE REDUCTION IN PARTICLE SIZE DURING CHEWING
Method
Symbol
Variable
Weighing
Qw (X)
The cumulative percentage undersize: the percentage of test food by weight that can pass a certain sieve aperture. Sieve aperture or particle size.
X
Curve-fitting equation (1)
X50 b
Linear interpolation
X20 X50 X80
Observation and counting
X20/X80 N
Curve-fitting equation (2)
X0 c, d
Calculation from equation (2)
N112
The aperture of a theoretical sieve for which QV = 50%, the median particle size. A measure of the extent to which the particles are equally sized. The apertures of theoretical sieves for which Q, = 20%, 50%, and 80%, respectively. A measure of the extent to which the particles are equally sized. The number of chewing strokes. The initial particle size Measures of the rate of reduction in median particle size. The number of chewing strokes for which X50 = 0.5 X0.
Equation (1): Q, (X) = 100 [1 - 2-xx-)b] Equation (2): X50(N) = (X0 c
Nd)/(Xo + c Nd)
Downloaded from jdr.sagepub.com at UNIV OF PITTSBURGH on December 18, 2014 For personal use only. No other uses without permission.
SLAGTER et al.
382
J Dent Res February 1992
TABLE 3 THE STATISTICAL FORMULAE USED FOR DETERMINATION OF THE INTRA- AND INTER-INDIVIDUAL VARIABILITY OF X20, X50, X80, c, d, AND N1/2
Formula
Purpose to estimate the intra-individual variance in the particle size distributions obtained from two sessions in terms of X20, X50, and X80 for each subject to estimate the intra-individual variance in the reduction in median particle size in terms of c, d, and N112 for each subject to calculate the relative intra-individual variation in each of the variables for each subject, expressed as a percentage
si, intra = (xi'l xi,2)) / 2
Sintra / [(x11 + x,) / 2] 100
X/2 +/ il i,2)
inter i== (( s2
x
(x
+x
)/2
)to estimate the inter-individual variance in each of the variables
n
for both the complete-denture wearers and the dentate subjects
n-1
Results. Fig. 1 shows the cumulative size distributions obtained from the pooled data of the denture-wearers and the dentate subjects after 20 chewing strokes. Estimates of X50 and b were not obtained, since the Rosin-Rammler equation did not adequately describe the data points. In Fig. 2, the curves representbestfits ofthe Rosin-Rammler equation for the pooled data. In Figs. 1 and 2, the mean, minimum, and maximum values for the individual estimates ofX20, X50, and X.0 are indicated as well. The group estimates of X50 and b and their corresponding 95% confidence intervals are given in Table 4. The dentate subjects and the denture-wearers differed markedly with respect to X20, X5 and X80, but overlapped with respect to the parameter b. All dentate subjects were able to comminute the test food particles to such an extent that after 20 strokes no particles were left on the upper sieve, with an aperture of 5.6 mm. Most denturewearers did not achieve a reduction in particle size to that extent, 100 -
Qw (%)
and even after 80 chewing strokes, up to 40% of the test food material by weight could be found on that sieve. In Table 5, the mean, median, minimum, and maximum percentages of the particles by weight that remained on the 5.6-mm sieve are given for each number of chewing strokes. The values for the parameters (X20, X50, X80, X20/X80, c, d, and did not differ significantly from a normal distribution N1/2) (Kolmogorov-Smirnov test). Differences in values of X20, X50, and X50 between denturewearers and dentate subjects were extremely significant (p < 0.0001). The mean values and standard deviations for both groups are given in Table 6. The inter-individual variance in the group of denturewearers was significantly larger than that in the group of dentate subjects (p < 0.05). One-tailed probability values of the differences in X53, X50, and X80 between 20,40, 60, and 80 chewing strokes were more significant in the group of denture-wearers (p < 0.005) than in the dentate group (p < 0.02). Between 20 and 80 chewing strokes, the reduction in X50 100 1
Qw (%) _
80 1
80 -
XI
L
0// 60 -
60 -
4X
LI
40
+
40
n
20
Li
0 10
0.1
x (mm)
Fig. 1-The cumulative percentage undersize by weight [Q, (X)] plotted
as a function of the sieve aperture (X) after 20 chewing strokes. The mean
values of X20, X50, and X80 for the complete-denture and the dentate groups are represented by an *. The minimum and maximum values of X20, X50, and X are indicated by a Dl for the complete-denture and a O for the dentate group. The curves represent the cumulative size distributions of each group: The dotted curve indicates the complete-denture and the dashed curve the dentate group.
Li
0 /,
20
-.0.1
.
( 1
10
x (mm)
Fig. 2-The cumulative percentage undersize by weight [Q, (X)] plotted
as a function of the sieve aperture (X) after 80 chewing strokes. The mean values of X20, X50, and X80 for the complete-denture and the dentate groups are represented by an *. The minimum and maximum values ofX20, X50, and are indicated by a En for the complete-denture and a O for the dentate Y.. group. The curves were obtained by curve-fitting the pooled data of each group, according to the Rosin-Rammler equation (1). The dotted curve indicates the complete-denture and the dashed curve the dentate group.
Downloaded from jdr.sagepub.com at UNIV OF PITTSBURGH on December 18, 2014 For personal use only. No other uses without permission.
383
FOOD COMMINUTION WITH COMPLETE DENTURES
Vol. 71 No. 2
TABLE 4 THE GROUP ESTIMATES AND 95% CONFIDENCE INTERVALS OF X50, b, c, AND d OBTAINED FROM THE POOLED DATA OF THE COMPLETE-DENTURE WEARERS AND THE DENTATE SUBJECTS AND THE CALCULATED VALUES OF N12 Number of chewing strokes
Complete-denture group
X50 (mm)
b
20 40
3.88
(3.67 - 4.09)
2.34
(1.89 - 2.79)
60
3.23
(3.03 - 3.44)
2.02
(1.65 - 2.39)
80
2.64
(2.47 - 2.82)
1.95
(1.62 - 2.28)
c (mm)
d
3983 (-3104 - 11,070)
-1.50 (-1.94 - -1.07)
X. (mm)
b
40
1.35 (1.29 - 1.42)
1.80
(1.59 - 2.01)
60
1.07 (1.04 - 1.09)
1.91
(1.81 - 2.01)
80
0.95 (0.93 - 0.97)
2.00
(1.89 - 2.11)
c (mm)
d
53.3 (34.9 - 71.7)
-0.91 (-1.00 - -0.81)
N1,2= 80 Number of chewing strokes Dentate group
20
N12 = 12 larger in the complete-denture wearers. The ratio X./X8 decreased significantly from 20 to 60 chewing strokes in the denture-wearers (p < 0.02), indicating that the particles were distributed over more different size classes as the number of chewing strokes increased. The ratio X,2X80 showed no significant change as a function ofthe number of chewing strokes in the dentate group. The intra-individual differences between sessions inX20, X50, and X80 were not significant for any number of chewing strokes (0.08 < p
