Vous êtes sur la page 1sur 7

Journal of Oral Rehabilitation

Journal of Oral Rehabilitation 2015 42; 537--543

A Randomised trial to investigate the erosive effect of hot


drinks
S. M. HOOPER*, E. L. MACDONALD*, M. ADDY*, M. ANTOGNOZZI†,
M . D A V I E S * & N . X . W E S T * *Restorative Dentistry, School of Oral and Dental Sciences, University of Bristol, Bristol,
and †Biomedical Research Unit in Cardiovascular Research Clinical Trials and Evaluation Unit, Bristol Heart Institute, University of
Bristol, Bristol, UK

SUMMARY Research on dental erosion has largely enamel specimen from 9 am to 5 pm for 15 days for
been undertaken at room temperature despite fruit each beverage. Measurements of enamel loss were
juice drinks often being consumed at elevated recorded after 5, 10 and 15 days by contact surface
temperatures in the United Kingdom, notably profilometry. The low erosive blackcurrant drink
during periods of convalesce. The aim of the study caused significantly less enamel loss (P < 005) than
was to evaluate the erosive potential of two fruit the commercially available conventional apple and
juices containing acidic non-alcoholic drinks at blackcurrant fruit juice drink and was not
elevated temperatures in situ on human enamel statistically significantly different to mineral water
after 5, 10 and 15 days. A commercially available at any of the time points in the study. Consuming
conventional apple and blackcurrant fruit juice the modified low erosive blackcurrant drink at an
drink was compared to a blackcurrant juice drink elevated temperature resulted in negligible enamel
modified to have low erosive potential, and mineral loss in situ, consistent with room temperature
water consumed at approximately 59 °C. Twenty- findings.
one healthy volunteers aged 18 or over participated KEYWORDS: dental erosion, temperature, hot drinks,
in a single-centre, single-blind (blinded to the fruit juice drinks, profilometry, in situ
investigator), three-treatment crossover study.
Subjects were randomised to a treatment sequence Accepted for publication 7 February 2015
using a Latin square design. Subjects wore upper
removable appliances containing one human

majority of research on erosion is performed at room


Background
or body temperatures.
The magnitude of variation in the susceptibility of The temperature at which the public consumes
individuals to erosion in relation to different behavio- acidic soft drinks varies. For example, in some drinks
ural, biological and chemical factors has been clearly such as Slush Puppie* , ice is a major component, so
demonstrated (1). Possible factors attributed to these the drink is consumed at a temperature just above
variations are age (2), saliva composition and flow as freezing. This contrasts sharply with concentrated fruit
well as pellicle formation (3) and abrasive influences juice drinks being diluted with boiling water and
of the tongue (4). It is known that factors such as the imbibed at high temperatures. In the United King-
nature, composition and frequency of acidic intake dom, hot fruit drinks are often part of cold and flu
are crucial to the degree of erosion (1). However,
other variables such as elevated temperature have
received limited attention (5), as interestingly, the *Slush Puppie Ltd., High Wycombe, UK.

© 2015 John Wiley & Sons Ltd doi: 10.1111/joor.12285


538 S . M . H O O P E R et al.

convalescence, with fruit-flavoured remedies such as juice (9%) drink concentrate [conventional fruit
Lemsip† and Beechams Powders‡ being commonly drink, CFD]**, which has similar fruit constituents to
consumed. However, the use of such hot remedies in the low erosive drink. Mineral water†† was selected as
other countries may not be as common. the negative control product.
Temperature has an important influence on the rate Polished flat human enamel samples were prepared
of chemical reactions as described by the Arrhenius as described previously by Hunter et al. (9) and a
equation (6). Previous in vitro studies have shown that baseline profile obtained with a tolerance limit of
temperature affects erosive potential (5), and Barbour 01 lm by contact profilometry (Surfometer SF200,
et al. (7) demonstrated that while enamel loss increased Planar Industrial‡‡). Specimens were taped to leave a
with increasing temperature (4–75 °C) following treat- 2-mm window of exposed enamel.
ment with a conventional fruit drink, enamel loss fol- NHS Research Ethical Committee approval was
lowing treatment with a drink modified to have a low obtained, and the study was designed, conducted and
erosive potential did not increase with temperature. This reported in accordance with the guidelines for Good
finding is pertinent to in situ investigation given that in Clinical Practice (10). Subjects gave informed consent to
the United Kingdom, the consumption of hot fruit- participate. All recruitment, randomisation, treatment
based drinks is relatively common. and measurements were carried out at a single site
This study aims to determine in situ whether a pre- (Dental Hospital). Identification numbers were allo-
viously documented low erosive blackcurrant drink cated in ascending numerical order based on the order
(8) retains its low erosive potential when imbibed at in which subjects were screened. Following screening,
elevated temperature, and, under these conditions, those who fulfilled the inclusion criteria of being
how erosive it is compared to a conventional soft fruit healthy adult volunteers and able to wear a dental appli-
juice drink and mineral water. ance were enrolled onto the study. An additional
screening procedure was undertaken to ensure that sub-
jects could consume drinks at 25 mL min 1 over a
Methods
10-min period at between 52 and 59 °C. Subjects were
This was a single-centre, single-blind (blinded to the then randomised to a treatment sequence using a Latin
person responsible for performing the contact profil- square design.
ometry measurements), three-way crossover study to To ensure that the drinks were supplied between 52
compare the erosive effect of two commercially avail- and 59 °C, 250 mL of concentrate test drink or mineral
able fruit juice drinks and mineral water, consumed water was dispensed into a thermos flask§§. Seven hun-
within the temperature range of 52–59 °C over dred and fifty millilitres of mineral water was added to
15 day periods. The objectives of the study were to an empty kettle, boiled and poured into the flask which
determine the erosive effect on tooth enamel, mea- was closed immediately and shaken well to achieve a
sured by contact profilometry, of the low erosive thoroughly mixed 1:4 diluted fruit juice drink or 100%
drink compared to a positive control (a conventional mineral water. The temperature of the diluted drinks at
soft fruit juice drink) and a negative control (mineral this point was 66 °C, and the drink was retained for up
water), all consumed at an elevated temperature, after to 1 h in the thermos flask. The 1:4 ratio used followed
5-,10- and 15-day treatment periods. the manufacturers’ recommendations for dilution of
The fruit juice drinks investigated were a low ero- concentrate for both fruit-flavoured drinks. For the
sive blackcurrant juice (7%) drink concentrate con- subjects, 250 mL portions of the prepared drinks were
taining calcium [low erosive, LE]§¶), noted for its low measured directly into 04 L thermos mugs§§ which
erosive potential at room temperature (8), and a no were closed by a cap to maximise control over tempera-
added sugar conventional apple and blackcurrant ture. The temperature of the dispensed drinks was


Reckitt Benckiser Group plc, Slough, UK. **Britvic Soft Drinks Ltd., Chelmsford, UK.
‡ ††
GlaxoSmithKline plc, Brentford, UK. Premier Waters Ltd., London, UK.
§ ‡‡
Lucozade Ribena Suntory Limited, Uxbridge, UK. Planar Products Ltd., Sunbury on Thames, UK.
¶ §§
formerly, GlaxoSmithKline UK Limited, Middlesex, UK. Thermos, BHL, Leeds, UK.

© 2015 John Wiley & Sons Ltd


IN SITU EROSION OF HOT DRINKS 539

approximately 59 °C. On arrival, subjects commenced from a previous study (11) where drinks were con-
supervised sipping of the drinks at a rate of approxi- sumed at room temperature and the erosion levels in
mately 25 mL min 1, for 10 min during which time the the positive control were quite variable. Using this
drinks cooled down by 5–7 °C. data and assuming two-sided paired t-tests would be
On study days (weekdays) between 9 am and performed for analysis at a significance level of 5%,
5 pm, subjects wore an intra-oral appliance contain- 18 subjects would provide over 80% power to detect
ing one enamel sample sited centrally and anteriorly a mean difference of 39 lm (standardised difference
in the palate. The appliance was removed for a 1-h of 075) between treatments. Sufficient subjects (to a
lunch period daily and for the duration of any tea, maximum of 22) were recruited to ensure at least 18
coffee or water consumption. There was no restriction evaluable subjects completed the trial successfully.
on food intake during the lunch break; however, Gross heterogeneity between erosive and non-
other beverages, food and/or acidic medication, were erosive formulations was expected from experience in
not permitted during the 9 am to 5 pm window. previous studies (12, 13). Accordingly, the primary
Volunteers visited the study site to consume study analysis was planned to consist of paired t-tests com-
drinks at 900, 1100, 1300 and 1500 h (all 30 min) paring selected pairs of study products for enamel loss
on each study day. The intra-oral appliances (containing at each time point and the differences to be estimated
the enamel samples) were soaked in a 02% w/v chlorh- with 95% confidence intervals. The primary time
exidine gluconate mouthrinse (Corsodylⶶ) for 3 min point was at 10 days. However, the assumptions
at the start of each treatment day. Subjects were required to conduct paired t-tests on the data were
required to wear their appliance prior to consumption of not satisfied; therefore, the Wilcoxon signed ranks test
the study drinks for pellicle formation. Those subjects was used, which was also a paired comparison, and
who dosed immediately prior to lunch were asked to median differences with 95% confidence intervals are
keep the appliance in their mouth after dosing to enable presented.
the saliva to neutralise the mouth (including the enamel
sample) before appliance removal. At the end of each
Results
treatment day, the intra-oral appliances were removed
from the mouth, soaked in chlorhexidine mouthrinse A total of 21 healthy adult volunteers were screened,
for 3 min and stored overnight in a moist pot. randomised and took part in the study. There were 12
On completion of days 5, 10 and 15, the enamel female and 9 male subjects with a mean age 34 years
samples were assessed for tissue loss by contact profil- and an age range of 21–60 years; 952% of the volun-
ometry. The tape was removed from either side of the teers were Caucasian. The intention-to-treat (ITT)
window of enamel exposed to the test drink, and population contained all 21 subjects. However,
samples were disinfected by placement in a mixture although the per protocol (PP) population also con-
of 05% chlorhexidine and 70% aqueous ethanol for tained all 21 subjects, data were not analysed from
a period of at least 20 min before contact profilometry one leg each of four subjects. Three subjects had data
measurements were performed. The mean of two eliminated due to concomitant use of medication con-
depth tracings across the unmasked area was recorded taining vitamin C. One further subject did not com-
and used for analysis. Before the samples were plete the final leg of the study. As a result, two CFD
replaced in the intra-oral appliances, they were disin- and two mineral water sets of data were lost from
fected in 05% chlorhexidine in a 70% aqueous etha- analysis per protocol (Fig. 1).
nol for 3 min and retaped. Baseline measurements for the enamel samples
were within the range 007 to +010 lm; differences
in enamel loss for the PP population are depicted in
Statistical and analytical methods
Fig. 2. When consumed hot, the low erosive blackcur-
An estimate of the standardised difference (mean dif- rant drink caused significantly less enamel loss than
ference/standard deviation of difference) was taken the conventional low sugar apple and blackcurrant
drink (positive control). The differences in enamel loss
among these products were statistically significant at
¶¶
GlaxoSmithKline, Weybridge, UK. all time points in the PP population (Table 1). No

© 2015 John Wiley & Sons Ltd


540 S . M . H O O P E R et al.

Fig. 1. Study design and progress of participants through the trial.

There were 56 adverse events recorded during the


study, and none were deemed by the investigators to
be related to the treatment. The 56 adverse events
were classified as 37 moderate and 19 mild events. All
study products were well tolerated. The adverse
events related to ailments reported by the partici-
pants; reported events were occasional headaches,
period pains, stomach cramps, minor head colds, a
minor chest infection and minor aches and pains from
Fig. 2. Median enamel loss (lm): change from baseline PP pop- limbs. No event was considered to be related to the
ulation (N = 21). CFD, conventional fruit drink; LE, low erosive
study medication. No action was required for any
drink.
reported event.
significant differences in enamel loss were found
between the low erosive drink and mineral water
Discussion
(negative control) at day 10 and all other time points
(P > 005). Statistically significant differences in Acidic soft drinks are known to cause the erosion of
enamel loss were detected between the conventional tooth tissue (11). However, studies that have investi-
drink and the mineral water (negative control) and gated the influence of temperature of soft drinks on
between the conventional drink and the low erosive erosion are limited (5, 7, 14). As in the United King-
drink at all time points. dom, it is not uncommon to drink hot fruit drinks

© 2015 John Wiley & Sons Ltd


IN SITU EROSION OF HOT DRINKS 541

Table 1. Median difference between treatments for enamel loss (lm) in the PP population (N = 21)

Day 5† Day 10† Day 15†

Median Median Median


Difference and Difference and Difference and
95% CI (lm) P-value 95% CI (lm) P-value 95% CI (lm) P-value

LE vs. CFD 008 (000, 026) 004* 017 (000, 072) 003* 035 (004, 101) 000*
CFD vs. Mineral 009 ( 027, 001) 002* 023 ( 080, 004) 000* 035 ( 103, 003) 000*
water
LE vs. Mineral 001 ( 004, 003) 059 002 ( 007, 003) 037 002 ( 006, 003) 053
water

LE, low erosive drink; CFD, conventional fruit drink.


Median difference is the median value of all possible treatment differences.
*Significant at the 5% level.

Positive values for the median difference indicate more enamel loss for the second formulation compared to the first formulation.

when suffering from a cold or flu, the extent of the dental erosion, and further work by these authors sug-
erosive effects of conventional fruit drinks and those gested links to age, gender and socio-economic influ-
modified to be of low erosive potential at elevated ence (16). Saliva flow rates also vary between
temperature is an important area of study. individuals; those with a dry mouth are more suscepti-
The results of this in situ study demonstrated that ble to acid erosion (3). Temperature fluctuations are
the low erosive beverage produced minimal erosion at known to exist intra-orally (17); hence, the position-
elevated temperatures. This finding supports those of ing of the enamel samples is also important. In the
a previous in vitro study in which it was demonstrated present study, samples were placed in the palatal
that a low erosive drink, similar to that used in the region as this is comfortable for the subject and fluid
present study, caused no significant tissue loss at tem- flow can more easily be ensured over this area. How-
peratures of up to 75 °C (7). The conventional fruit ever, it is likely that more erosion would be demon-
drink used in the comparative in vitro study was also strated if samples were placed more anteriorly in the
similar to that used in the present study, and in vitro, upper incisor region, not least because during drinking
the conventional drink was shown to cause consider- this is often the first contact for the acid on tooth.
able tooth tissue loss at elevated temperatures in the The low erosive fruit drink used in the present
comparative study. Interestingly, in the present study, study contains elevated levels of calcium as compared
the conventional control drink did not cause as much to conventional fruit drinks, and Barbour et al. (7)
tissue loss at elevated temperature in situ as might demonstrated that compared to the control drink, the
have been expected from the findings of the in vitro low erosive fruit drink also had a significantly higher
study, although it did produce significantly more ero- pH, a lower titratable acidity and a higher degree of
sion than the low erosive drink and mineral water at saturation with respect to hydroxyapatite. Further, it
all time points investigated. was demonstrated that these differences were main-
Differences in the magnitude of enamel softening or tained with increasing temperature. However, given
tooth wear between in vitro and in situ studies reflect the similarity of the drinks investigated in the two
differences in the protocols employed such as exposure studies, these factors are likely to be responsible for
time to acid challenge and the flow rate of the acid the differences observed between the erosive poten-
challenge. Faster flow rates have been shown to cause tials of the drinks as tested in situ. By contrast, it has
more tissue loss (15). In addition, human saliva which been demonstrated that calcium phosphate added to
is known to protect against acid erosion (3) is also orange juice showed a preventive effect against ero-
often absent in in vitro protocols. Variability in in situ sion even though pH and buffering capacity were not
studies occurs due to differences in the susceptibility of affected (18).
subjects to erosive challenge. A longitudinal study by The efficacy of added calcium in minimising erosive
El Aidi et al. (2), has shown differing susceptibility to potential at elevated temperatures suggests that it

© 2015 John Wiley & Sons Ltd


542 S . M . H O O P E R et al.

could be added to other drinks designed to be con- 5. West NX, Hughes JA, Addy M. Erosion of dentine and
sumed at elevated temperatures such as cold and flu enamel in vitro by dietary acids: the effect of temperature,
acid character, concentration and exposure time. J Oral
remedies providing it did not interfere with the active
Rehabil. 2000;27:875–880.
ingredients. One disadvantage of calcium is the nega- 6. Connors KA. Chemical kinetics: the study of reaction rates
tive affect that it can have on taste (19). Several alter- in solution. New Jersey, USA: Wiley-Blackwell; 1990:187–
natives that might reduce the erosive capacity of 188.
drinks have been tested in vitro including various food 7. Barbour ME, Finke M, Parker DM, Hughes JA, Allen GC,
polymers, casein phosphopeptides – amorphous casein Addy M. The relationship between enamel softening and
erosion caused by soft drinks at a range of temperatures. J
phosphate [CPP-ACP], and nano-sized hydroxyapatite,
Dent. 2006;34:207–213.
and these have been shown to have varying degrees 8. Hughes JA, West NX, Parker DM, Newcombe RG, Addy M.
of efficacy (20–22). In situ, it has been demonstrated Development and evaluation of a low erosive blackcurrant
that CPP-ACP can remineralise enamel softened by juice drink 3. Final drink and concentrate, formulae com-
exposure to cola (23), and one study has demon- parisons in situ and overview of the concept. J Dent.
1999;27:345–350.
strated that food polymers can reduce the erosive
9. Hunter ML, Hughes JA, Parker DM, West NX, Newcombe
potential of drinks (24). The number of in situ studies RG, Addy M. Development of low erosive carbonated fruit
is very limited, and more work is needed in this field drinks. 1. Evaluation of two experimental orange drinks
of research. in vitro and in situ. J Dent. 2003;31:253–260.
10. ICH. Topic E6 (R1) Guidelines for Good Clinical Practice. CPMP
2002 July; ICH 135 95. Available at: http://www.ema.euro-
Conclusion pa.eu/docs/en_GB/document_library/Scientific_guideline/
2009/09/WC500002874.pdf, accessed 23 February 2015.
Consuming the modified low erosive blackcurrant 11. Hooper S, West NX, Sharif N, Smith S, North M, De’Ath J
drink at an elevated temperature resulted in negligible et al. A comparison of enamel erosion by a new sports drink
enamel loss in situ, consistent with room temperature compared to two proprietary products: a controlled, cross-
findings. Hence, calcium modification should be over study in situ. J Dent. 2004;32:541–545.
12. Hughes JA, West NX, Parker DM, Newcombe RG, Addy M.
considered for acidic fruit drinks that are designed to
Development and evaluation of a low erosive blackcurrant
be consumed at elevated temperatures. juice drink in vitro and in situ 1. Comparison with orange
juice. J Dent. 1999;27:285–289.
13. West NX, Hughes JA, Parker DM, Newcombe RG, Addy M.
Acknowledgments Development and evaluation of a low erosive blackcurrant
juice drink 2. Comparison with a conventional blackcurrant
This study was approved by the Central and South
juice drink and orange juice. J Dent. 1999;27:341–344.
Bristol Research Ethics Committee (E5875). The 14. Amaechi BT, Higham SM, Edgar WM. Factors influencing
research was supported by GlaxoSmithKline, the development of dental erosion in vitro: enamel type,
Weybridge, UK (NHCMA0303). temperature and exposure time. J Oral Rehabil.
1999;26:624–630.
15. Eisenburger M, Addy M. Influence of liquid temperature
Disclosure and flow rate on enamel erosion and surface softening. J
Oral Rehabil. 2003;30:1076–1080.
The authors declare there are no conflicts of interests. 16. El Aidi H, Bronkhorst EM, Huysmans MC, Truin GJ.
Dynamics of tooth erosion in adolescents: a 3-year longitu-
dinal study. J Dent. 2010;38:131–137.
References 17. Airoldi G, Riva G, Vanelli M, Filippi V, Garattini G. Oral envi-
ronment temperature changes induced by cold/hot liquid
1. Magalhᾶes AC, Wiegand A, Rios D, Hon orio HM, Buzalaf
intake. Am J Orthod Dentofac Orthop. 1997;112:58–63.
MAR. Insights into preventive measures for dental erosion.
18. Larsen MJ, Nyvad B. Enamel erosion by some soft drinks
J Appl Oral Sci. 2009;17:75–86.
and orange juices relative to their pH, buffering effect and
2. El Aidi H, Bronkhorst EM, Truin GJ. A longitudinal study of
contents of calcium phosphate. Caries Res. 1999;33:81–87.
tooth erosion in adolescents. J Dent Res. 2008;87:731–735.
19. Cochrane NJ, Yuan Y, Walker GD, Shen P, Chang CH, Rey-
3. Buzalaf MAR, Hannas AR, Kato MT. Saliva and dental ero-
nolds C et al. Erosive potential of sports beverages. Aust
sion. J Appl Oral Sci. 2012;20:493–502.
Dent J. 2012;57:359–364.
4. Gregg T, Mace S, West NX, Addy M. A study in vitro of the
20. Ferrazzano GF, Coda M, Cantile T, Sangianantoni G, Ingeni-
abrasive effect of the tongue on enamel and dentine soft-
to A. SEM investigation on casein phosphopeptides capabil-
ened by acid erosion. Caries Res. 2004;38:557–560.

© 2015 John Wiley & Sons Ltd


IN SITU EROSION OF HOT DRINKS 543

ity in contrasting cola drinks enamel erosion: an in vitro with 900 ppm fluoride on eroded human enamel: an in situ
preliminary study. Eur J Paediatr Dent. 2012;13:285–288. study. Arch Oral Biol. 2010;55:541–544.
21. Min JH, Kwon HK, Kim BI. The addition of nano-sized 24. Hooper S, Hughes J, Parker D, Finke M, Newcombe RG,
hydroxyapatite to a sports drink to inhibit dental erosion – Addy M et al. A clinical study in situ to assess the effect of a
in vitro study using bovine enamel. J Dent. 2011;39:629–635. food approved polymer on the erosion potential of drinks. J
22. Barbour ME, Shellis RP, Parker DM, Allen GC, Addy M. An Dent. 2007;35:541–546.
investigation of some food-approved polymers as agents to
inhibit hydroxyapatite dissolution. Eur J Oral Sci. Correspondence: Susan M Hooper, School of Oral and Dental
2005;113:457–461. Science, Bristol Dental Hospital, Lower Maudlin Street, Bristol BS1
23. Srinivasan N, Kavitha M, Loganathan SC. Comparison of 2LY, UK. E-mail: susan.hooper@bristol.ac.uk
the remineralization potential of CPP-ACP and CPP-ACP

© 2015 John Wiley & Sons Ltd