|
Sign In to gain access to subscriptions and/or personal tools.
|
Immediate Erosive Potential of Cola Drinks and Orange Juices
T. Jensdottir1,*,2,
P. Holbrook3,
B. Nauntofte1,
C. Buchwald4 and
A. Bardow1
1 Department of Oral Medicine, School of Dentistry, University of Copenhagen, Nørre Alle 20, 2200 Copenhagen N, Denmark;
2 Toms Group A/S, Ballerup, Denmark;
3 Faculty of Odontology, University of Iceland, Vatnsmyrarvegur 16, 101 Reykjavik, Iceland; and
4 Department of Otolaryngology, Head and Neck Surgery, Rigshospitalet, Copenhagen, Denmark
Correspondence: * corresponding author, tje{at}odont.ku.dk
 |
ABSTRACT
|
|---|
Little is known about the erosive potential of soft drinks within the first minutes of exposure to teeth, and about the potentially protective role of salivary proteins. We hypothesized that the erosive potential is determined primarily by pH and decreases in the presence of salivary proteins. To investigate this, we first added uncoated hydroxyapatite crystals and, second, salivary-protein-coated hydroxyapatite crystals to 20 commercially available cola drinks and orange juices simultaneously, with pH recordings every 15 sec for 3 min. The amount of apatite lost per liter of soft drink per sec was calculated from titratable acidity values to each pH obtained by crystal addition. The erosive potential within the first minutes of exposure was determined solely by the pH of the drink, and the erosive potential was ten-fold higher in cola drinks compared with juices. However, salivary proteins reduced the erosive potential of cola drinks by up to 50%.
Key Words: erosion soft drinks human salivary proteins hydroxyapatite crystals
 |
INTRODUCTION
|
|---|
Dental erosion is the chemical wear of the dental hard tissue without the involvement of bacteria (Eccles, 1979). The prevalence of dental erosion is increasing (Arnadottir et al., 2003; Nunn et al., 2003), and soft drink consumption is recognized as one of the main risk factors (Johansson et al., 1997). Clinical studies have found carbonated drinks, especially carbonated cola drinks, to be associated with erosion, most likely due to their low pH (Johansson et al., 1997; Jensdottir et al., 2004). However, in vitro studies have shown that fruit juices may also be potentially erosive, due to their high content of titratable acid (Lussi et al., 1995; Larsen and Nyvad, 1999; Jensdottir et al., 2005a). We speculate that the dynamics of the erosive potential within the first seconds and minutes of exposure may be critical, since the bulk of a soft drink stays in the mouth for only seconds before being swallowed. After swallowing occurs, the residual amount of liquid in the mouth will be reduced to less than 1 mL (Lagerlöf and Dawes, 1984), leaving only a limited amount of drink in contact with the teeth. In healthy individuals, and with reduced rate in dry-mouth patients, the soft drink is mixed with saliva, which then will re-establish its super-saturation level with respect to hydroxyapatite, due to acid clearance (Bashir and Lagerlöf, 1996) and salivary buffering capacity (Jensdottir et al., 2005b).
Equally important in the mouth are the protective effects of the salivary proteins that also may influence the erosive potential of soft drinks (Zahradnik et al., 1976; Meurman and Frank, 1991). The aim of this study was to determine the erosive effects of soft drinks within the first minutes of exposure, and the protective effects of salivary proteins. We hypothesized that the erosive potential of acidic drinks within the first minutes of exposure is closely related to clinical findings showing that soft drink pH is the most important factor for dental erosion, that, due to their low pH, cola drinks are, initially, considerably more erosive than orange juices, and that human salivary proteins may reduce the erosive potential during the acidic challenge.
 |
MATERIALS & METHODS
|
|---|
Ten orange juices and 10 cola drinks were tested. The cola drinks were: Coca Cola, Cola light, Pepsi Cola (all in plastic and glass), Cola light lemon, Pepsi Max, Pepsi Twist, and a local discount Cola drink. Orange juices were: Capri-Sonne Orange, Sun Top, Rynkeby with sweet oranges, sour oranges, and organic oranges, three orange juices made from concentrate, and two made from fresh orange juice, the latter five produced by local supermarket chains. Initially, the pH was recorded (pH0) after 50 mL of the drink was titrated with 1 M NaOH to a pH above 5.5 (Fig. 1A ). Then, a 50-mg quantity of freeze-dried hydroxyapatite (HAp) crystals was added to new 50-mL samples of each drink while the pH of the drinks was recorded at 15-second intervals for 3 min (pH1–pH12) and then finally 30 min (pH13) after HAp addition (Fig. 1B ). Within the pH range recorded (pH 2.5–4.5), HAp mainly dissolves as: Ca10(PO4)6(OH)2 10Ca2+ + 6H2PO4 + 2H2O where the released PO43– and OH– ions combine with 14 hydrogen ions, thereby changing the pH in an alkaline direction. The magnitude of this pH rise depends on the amount of titratable acid in the drink. Therefore, the volume of base (µL 1 M NaOH) needed to reach each pH value obtained by HAp addition (pH1–pH13) was determined from the titration curve (Fig. 1C ). Thus, titration with 14 µL of 1 M NaOH represents the loss of 14 µg of hydrogen ions in the drink, corresponding to dissolution of 1005 µg of HAp (MW 1005) crystals due to the stoichiometry of the reaction.

View larger version (15K):
[in this window]
[in a new window]
|
Figure 1. Experimental set-up. (A) pH measurements (pH0) and titration of soft drink with 1 M NaOH to a pH above 5.5 (N = 20). From these data, the titratable acidity was determined in µL 1 M NaOH. (B) Addition of 50 mg of pure hydroxyapatite (HAp) crystals to the drink (N = 20), under constant stirring, and pH recordings every 15 sec (pH1–pH12). After the initial recordings, the drink was left with constant stirring until the last pH recording was obtained after 30 min (pH13). (C) Titratable acidity (from Fig. 1A) up to each pH value obtained over time in response to HAp dissolution in the drinks (from Fig. 1B). The amount (µL) of base (1 M NaOH) needed to obtain each pH value induced by HAp addition was used to calculate the amount of HAp crystals dissolved (mg) per sec per liter soft drink (N = 20). In (C), iEP denotes the initial erosive potential (i.e., during the first minutes of exposure), and eEP denotes the end erosive potential (i.e., after 30 min).
|
|
Initial and End Erosive Potential
The initial erosive potential was determined as the erosive potential within the first 3 min of exposure of HAp to the drink and was recorded as the slope of a curve obtained by linear regression showing the HAp dissolved (in mg) per liter of drink as a function of time (in sec). When the slope of this curve was non-linear, data points were eliminated successively from pH12 and backward. In all cases, data points obtained within the first minute were kept in the analysis. The end erosive potential was recorded as mg HAp dissolved per liter of drink at pH13.
Effects of Human Salivary Proteins
The combined procedure was repeated with 50 mg of HAp crystals pre-treated with 5 mg of human salivary proteins. The proteins were dialyzed and lyophilized from a pool of 1 liter unstimulated and stimulated clarified whole saliva collected from 100 healthy dental students (upon ethical approval and informed consent). SDS-PAGE revealed that all major salivary proteins were represented in the pool (Schwartz et al., 1995). HAp crystals were coated with the proteins for 24 hrs, resembling the time between daily toothbrushings, at a temperature of 5°C, to prevent denaturation and bacterial growth, in a volume of 2 mL Millipore water at pH 6.5. This pH value allowed for the normal physiological functionality of the proteins. After the crystals were coated, they and the remaining excess protein in the 2-mL solution were lyophilized and added directly to the drink in a manner similar to that for the non-coated crystals. All experiments were carried out at room temperature and were repeated at least three times.
Statistics
Statistical analyses were done with Excel and the R statistical program (R Development Core Team, 2004). Differences between juices and cola drinks were analyzed by Wilcoxons rank-sum test and correlations with Spearmans rank correlation analysis (rs). For determination of the initial erosive potential (Fig. 2 ), the best linear relationships between HAp dissolution and exposure time to the drinks were obtained by linear regression analysis judged from the R-squared values obtained. In Fig. 3 , the curves were exponentially fitted. The level of significance was set at p < 0.05.

View larger version (25K):
[in this window]
[in a new window]
|
Figure 2. Erosive potential over time. (A) Erosive potential of 5 carbonated cola drinks and 5 orange juices selected as representatives of their groups for the first 3 min upon exposure to HAp crystals. (B) Erosive potential of the same drinks over the whole 30-minute test period. All drinks in the study (N = 20) were tested at least 3 times, with a mean percent relative standard deviation between repetitions of 24% for the juices and 23% for the colas. As shown, the sequence of the drinks changed, so that some of the orange juices became considerably erosive with time. The 5 carbonated cola drinks (1–5) and 5 orange juices (6–10) in Fig. 2 are (1) Coca Cola light, (2) Pepsi Max, (3) Coca Cola, (4) Pepsi Cola, (5) Coca Cola light with lemon, (6) Capri-Sonne Orange, (7) Sun Top, (8) Rynkeby with sour oranges, (9) Rynkeby with organic oranges, and (10) Rynkeby with sweet oranges.
|
|

View larger version (12K):
[in this window]
[in a new window]
|
Figure 3. Erosive potential and soft drink pH. (A) Relationship between the initial erosive potential (i.e., erosive potential during the first minutes of exposure) and the pH (open circles) for all drinks (N = 20). As shown, the initial erosive potential was almost a logarithmic function of the pH, increasing ten-fold for each one-unit decrease in pH. The erosive potential of the drinks was reduced, more so in drinks with low pH values and high initial erosive potential, when the HAp crystals were coated with human salivary proteins (bold circles), illustrated by the gray area between the lines (N = 20). (B) Corresponding relationship between the end erosive potential (i.e., erosive potential after 30 min) and the pH in the drinks (N = 20). Only a very limited effect was seen from the human salivary protein on the end erosive potential of the drinks, as illustrated by the reduction in size of the grey area compared with Fig. 3A (N = 20).
|
|
 |
RESULTS
|
|---|
The mean pH of the cola drinks was more than one pH unit lower than the mean pH of the orange juices (p < 0.001) (Table ). Despite the higher pH in the orange juices, their titratable acidity values to reach pH 5.5 were nearly five-fold higher, on average, than in the cola drinks (p < 0.001). Nonetheless, the cola drinks clearly had a higher erosive potential within the first minute of exposure to HAp crystals than did the orange juices (Fig. 2A ). Thus, the initial erosive potential in the cola drinks was more than ten-fold higher (p < 0.001) than in the orange juices (Table ), and the initial erosive potential was high in all cola drinks. In contrast, considerable differences were obtained among the orange juices, ranging from an initial erosive potential of around half that of the cola drinks to no sign of erosive potential (Fig. 2A ).
The erosive potential changed over time, more so in the cola drinks (Fig. 2B ). Thus, after the first 3 min, the erosive potential in the cola drinks slowed more than 40-fold, whereas the erosive potential slowed only less than three-fold in the orange juices. The slowing of the erosive potential was not due to lack of HAp crystals, since only 36% of the crystals in the cola drinks and only 6% in the juices were lost after 3 min, but due to increased pH. A significant negative correlation was obtained between the slowing of the erosive potential (initial erosive potential/erosive potential from 3–30 min) and the titratable acid values for the drinks (rs = 0.83; p < 0.001). Thus, the juices were able to maintain low pH values throughout the study period, with a pH rise of only 0.13 ± 0.14 unit compared with 1.79 ± 0.45 unit in the cola drinks (p < 0.001), resulting in a more constant erosive potential. These dynamics led to the finding that, among the orange juices, which, as a group, were only slightly erosive for the first 3 min, two drinks became the third and fourth most erosive drinks after 30 min (Fig. 2B ).
Effects of pH
The initial erosive potential was almost an exponential function (R-squared 0.81; p < 0.001) of the pH of the drinks, increasing nearly ten-fold for each time the pH lowered one unit (Fig. 3A ). Surprisingly, a negative relationship was obtained between the titratable acidity to pH 5.5 and initial erosive potential (not shown). When the data were reviewed, it became clear that this finding was due to all juices having higher titratable acid values than the cola drinks, and, at the same time, a lower initial erosive potential. The relation between pH and the end erosive potential after 30 min was quite different from that for the initial erosive potential (Fig. 3B ). Thus, the end erosive potential increased only around two-fold for every unit the pH was lowered. This shows that, upon prolonged exposure time to a limited volume of soft drink, other factors such as titratable acid also became important for the erosive potential.
Effects of Salivary Proteins
Prior to all experiments, the effect of adding 5 mg of proteins to 50 mL of drink was tested. For all drinks, protein addition had no, or only a negligible, effect on the pH and titratable acidity of the drink. Therefore, we were able to test the effect of adding the same amount of proteins now delivered with the HAp crystals to the drinks. The pre-treatment of HAp crystals with salivary proteins reduced the initial erosive potential by 50% at pH values near 2.5 (Fig. 3A ). However, at pH values above 3.5, no protective effect was obtained from the proteins. Thus, the protective effect of the proteins on HAp crystals (i.e., the relative reduction in initial erosive potential with salivary proteins) was significantly negatively correlated with pH (rs = –0.47; p < 0.05) and significantly positively correlated with the initial erosive potential (rs = 0.65; p < 0.01). As a consequence of this relationship, the protective effect was higher in cola drinks than in juices on a group basis (p < 0.05). Interestingly, only a very limited effect of the proteins was obtained on the end erosive potential after 30 min, and this effect was not dependent on pH (Fig. 3B ).
 |
DISCUSSION
|
|---|
To predict the erosive potential of a soft drink, the method used should simulate what happens in vivo when the drink enters the mouth. Theoretically, this must be dependent upon the immediate effect of the drink on the tooth surface, the time it takes to clear the drink from the mouth (Bashir and Lagerlöf, 1996), the drinking method (Johansson et al., 2004), the protective effect of saliva (Zahradnik et al., 1976; Meurman and Frank et al., 1991; Jensdottir et al., 2005b), and the amount of residual drink after swallowing. Although the HAp method used in this study is purely experimental, creating a situation with a large crystal surface area in contact with the drink, it offers the possibility for mineral dissolution to be monitored almost instantaneously upon contact with the drink tested.
Cola drinks had more than ten-fold higher erosive potential than orange juices within the first minutes after exposure. This high erosive potential corresponded well to the pH of the cola drinks, which was around one unit lower than that of the juices. Thus, within the first minutes, the erosive potential was nearly an exponential function of pH in both cola drinks and orange juice, as would be expected due to the logarithmic nature of the pH scale. These findings are in agreement with those of Larsen and Nyvad (1999), who found a similar exponential relation between soft drink pH and erosive potential on teeth. However, in their study, the titratable acidity was also found to have an effect on the erosive potential, a finding that has been supported by several other studies (Lussi et al., 1995; Edwards et al., 1999; Jensdottir et al., 2005a). Nevertheless, according to our in vitro findings, we speculate that the titratable acidity is not related to the erosive potential from the time the drink meets the tooth surfaces until the first swallow occurs. The titratable acidity may, however, become important later on, when some of the drink is kept in contact with teeth. Such a situation could occur in dry-mouth patients with low salivary flow rates (Fox et al., 1987) and, consequently, slow oral clearance (Dawes, 1983), with drinks that, due to their physical characteristics, tend to attach to the teeth for a long period of time (Ireland et al., 1995), and in patients with special drinking habits (Ireland et al., 1995; Johansson et al., 2004).
In the mouth, tooth surfaces are covered with the acquired pellicle, comprising many of the proteins present in saliva (Lendenmann et al., 2000), and this pellicle has been shown to protect tooth surfaces against erosion (Meurman and Frank, 1991; Amaechi et al., 1999; Nekrashevych and Stosser, 2003). In this study, we showed that the protective effects of salivary proteins increased with increasing erosive potential within the first minutes of exposure to the drinks. Within this time, the proteins halved the erosive potential of cola drinks with a low pH, while only limited effect was found with the orange juices that had higher pH values and thus lower erosive potential. We speculate that these findings are due to the relationship between the speed of desorption of proteins from the crystals and the speed of the erosive challenge. Thus, if the erosive speed (initial erosive potential) exceeded the speed of desorption of proteins from the crystals, a protective effect was obtained, and vice versa. We assume that when the proteins were washed of the crystals, their protective effect ceased, which explains the very limited effects of the proteins on the erosive potential after 30 min (end erosive potential). Thus, in the mouth, after exhibiting its protective effects, the protein coating must be renewed to withstand a new acidic challenge. However, renewal of the protein coating may take considerable time (Zahradnik et al., 1976; Nieuw Amerongen, 1987), and this time may be a critical factor in individuals who tend to sip soft drinks throughout the day, which partly explains their high incidence of erosion (Johansson et al., 2004). Another likely explanation for the relationship between the protective effects of the proteins and the erosive potential of the drinks may be protein denaturation, which could have occurred in drinks with low pH values, thereby increasing the viscosity of the protein coating on the crystals (Holma and Hegg, 1989) and adding to the protective effects.
In conclusion, this study shows that the erosive potential of soft drinks within the first minutes of exposure is solely dependent on the pH of the drinks. Furthermore, it is ten-fold higher in cola drinks when compared with orange juices. However, in cola drinks, human salivary proteins may reduce the erosive potential up to 50%.
 |
ACKNOWLEDGMENTS
|
|---|
We are grateful for the donation of HAp crystals from Dr. J. Christoffersen and Dr. M.R. Christoffersen (Uni-Crystals, Vallensbaek, Denmark). The Danish Ministry of Science, Technology and Innovation supported the study. Preliminary reports were presented at the IADR meetings in Gothenburg (2003) and Honolulu (2004).
Received for publication April 29, 2005.
Revision received October 26, 2005.
Accepted for publication November 1, 2005.
 |
REFERENCES
|
|---|
- Amaechi BT, Higham SM, Edgar WM, Milosevic A (1999). Thickness of acquired salivary pellicle as a determinant of the sites of dental erosion. J Dent Res 78:1821–1828.
- Arnadottir IB, Sæmundsson SR, Holbrook WP (2003). Dental erosion in Icelandic teenagers in relation to dietary and lifestyle factors. Acta Odontol Scand 61:25–28.[Medline]
[Order article via Infotrieve]
- Bashir E, Lagerlöf F (1996). Effect of citric acid clearance on the saturation with respect to hydroxyapatite in saliva. Caries Res 30:213–217.[Medline]
[Order article via Infotrieve]
- Dawes C (1983). A mathematical model of salivary clearance of sugar from the oral cavity. Caries Res 17:321–334.[Medline]
[Order article via Infotrieve]
- Eccles JD (1979). Dental erosion of non-industrial origin. A clinical survey and classification. J Prosthet Dent 42:649–653.[Medline]
[Order article via Infotrieve]
- Edwards M, Creanor SL, Foye RH, Gilmour WH (1999). Buffering capacities of soft drinks: the potential influence on dental erosion. J Oral Rehabil 26:923–927.[Medline]
[Order article via Infotrieve]
- Fox PC, Busch KA, Baum BJ (1987). Subjective reports of xerostomia and objective measures of salivary gland performance. J Am Dent Assoc 115:581–584.[Abstract]
- Holma B, Hegg PO (1989). pH- and protein-dependent buffer capacity and viscosity of respiratory mucus. Their interrelationships and influence on health. Sci Total Environ 84:71–82.[CrossRef][Medline]
[Order article via Infotrieve]
- Ireland AJ, McGuinness N, Sherriff M (1995). An investigation into the ability of soft drinks to adhere to enamel. Caries Res 29:470–476.[Medline]
[Order article via Infotrieve]
- Jensdottir T, Arnadottir IB, Thorsdottir I, Bardow A, Gudmundsson K, Theodors A, et al. (2004). Relationship between dental erosion, soft drink consumption, and gastroesophageal reflux among Icelanders. Clin Oral Inv 8:91–96.
- Jensdottir T, Bardow A, Holbrook WP (2005a). Properties and modification of soft drinks in relation to their erosive potential in vitro. J Dent 33:569–575.[Medline]
[Order article via Infotrieve]
- Jensdottir T, Nauntofte B, Buchwald C, Bardow A (2005b). Effects of sucking acidic candy on whole mouth saliva composition. Caries Res 39:468–474.[CrossRef][Medline]
[Order article via Infotrieve]
- Johansson AK, Johansson A, Birkhed D, Omar R, Baghdadi S, Khan N, Carlsson GE (1997). Dental erosion associated with soft-drink consumption in young Saudi men. Acta Odontol Scand 55:390–397.[Medline]
[Order article via Infotrieve]
- Johansson AK, Lingström P, Imfeld T, Birkhed D (2004). Influence of drinking method on tooth-surface pH in relation to dental erosion. Eur J Oral Sci 112:484–489.[Medline]
[Order article via Infotrieve]
- Lagerlöf F, Dawes C (1984). The volume of saliva in the mouth before and after swallowing. J Dent Res 63:618–621.
- Larsen MJ, Nyvad B (1999). Enamel erosion by some soft drinks and orange juices relative to their pH, buffering effect and contents of calcium phosphate. Caries Res 33:81–87.[Medline]
[Order article via Infotrieve]
- Lendenmann U, Grogan J, Oppenheim FG (2000). Saliva and dental pellicle—a review. Adv Dent Res 14:22–28.[Abstract/Free Full Text]
- Lussi A, Jaeggi T, Jaeggi-Scharer S (1995). Prediction of the erosive potential of some beverages. Caries Res 29:349–354.[Medline]
[Order article via Infotrieve]
- Meurman JH, Frank RM (1991). Scanning electron microscopic study of the effect of salivary pellicle on enamel erosion. Caries Res 25:1–6.[Medline]
[Order article via Infotrieve]
- Nekrashevych Y, Stosser L (2003). Protective influence of experimentally formed salivary pellicle on enamel erosion. An in vitro study. Caries Res 37:225–231.[Medline]
[Order article via Infotrieve]
- Nieuw Amerongen AV, Oderkerk CH, Driessen AA (1987). Role of mucins from human whole saliva in the protection of tooth enamel against demineralization in vitro. Caries Res 21:297–309.[Medline]
[Order article via Infotrieve]
- Nunn JH, Gordon PH, Morris AJ, Pine CM, Walker A (2003). Dental erosion—changing prevalence? A review of British National childrens surveys. Int J Paed Dent 13:98–105.
- R Development Core Team (2004). R: a language and environment for statistical computing, Vienna, Austria. ISBN 3-900051-00-3, URL http://www.R-project.org.
- Schwartz SS, Zhu WX, Sreebny LM (1995). Sodium dodecyl sulphate-polyacrylamide gel electrophoresis of human whole saliva. Arch Oral Biol 40:949–958.[CrossRef][Medline]
[Order article via Infotrieve]
- Zahradnik RT, Moreno EC, Burke EJ (1976). Effect of salivary pellicle on enamel subsurface demineralization in vitro. J Dent Res 55:664–670.
Journal of Dental Research, Vol. 85, No. 3,
226-230 (2006)
DOI: 10.1177/154405910608500304

CiteULike Complore Connotea Del.icio.us Digg Reddit Technorati Twitter What's this?
This article has been cited by other articles:

|
 |

|
 |
 
S. N. Wagoner, T. A. Marshall, F. Qian, and J. S. Wefel
In Vitro Enamel Erosion Associated With Commercially Available Original-Flavor and Sour Versions of Candies
J Am Dent Assoc,
July 1, 2009;
140(7):
906 - 913.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
W.P. Holbrook, J. Furuholm, K. Gudmundsson, A. Theodors, and J.H. Meurman
Gastric Reflux is a Significant Causative Factor of Tooth Erosion
Journal of Dental Research,
May 1, 2009;
88(5):
422 - 426.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
R. E. Davis, T. A. Marshall, F. Qian, J. J. Warren, and J. S. Wefel
In Vitro Protection Against Dental Erosion Afforded by Commercially Available, Calcium-Fortified 100 Percent Juices
J Am Dent Assoc,
December 1, 2007;
138(12):
1593 - 1598.
[Abstract]
[Full Text]
[PDF]
|
 |
|
|
|