Cariogenic potential of oral nutritional supplements measured by intraoral plaque pH telemetry.

BACKGROUND & AIMS
Oral nutritional supplements (ONS) administered to malnourished elders and other patients contain high-levels of carbohydrates that could be a potential risk factor for dental caries. This study aimed to evaluate the cariogenic potentials of ONS using intraoral plaque telemetry.


METHODS
Ten ONS were tested on five healthy volunteers (mean age: 76.8 ± 9.15 years). Participants were requested to refrain from performing oral hygiene 3-7 days prior to testing. The pH-value below the dental plaque on the tooth was measured while the ONS was being consumed. After neutralizing the participant's saliva, a control solution (10% sucrose) was administered and telemetry measurements were repeated. Mean relative cariogenicity (RC) was calculated for each ONS. ANOVA and post hoc tests were used for statistical analyses (p < 0.05).


RESULTS
All ten ONS were potentially cariogenic on enamel with an overall RC of 0.519 ± 0.35 (Range: Min = 0.31 ± 0.16; Max = 1.00 ± 0.34). RC differed significantly between the ONS (p = 0.002). RC was lower in ONS that contained high-protein (p = 0.018). RC was not influenced by other factors such as readily consumable (p = 0.102), flavor (p = 0.869), consistency (p = 0.126), fiber containing (p = 0.134), style (p = 0.112), and age of plaque (p = 0.339).


CONCLUSIONS
The ONS administered to elders and malnourished patients are potentially cariogenic. It is imperative that the administration of ONS must be based on individual needs to potentiate a maximum benefit. Wherever possible, an attempt to limit the use of high-carbohydrate containing ONS must be practiced along with the adoption of suitable preventive measures to arrest the development and progression of caries.


Introduction
Oral nutritional supplements (ONS) are administered to malnourished patients as well as frail, care-dependent elders in different health care settings to reduce the burden of malnutrition, strengthen function, and improve their quality of life [1]. These supplements usually contain high levels of carbohydrates, amongst other essential constituents, which counter malnutrition. The current global population demographic trends suggest that the elderly population is increasing [2,3], and in developed nations teeth are being retained to an advanced age [4e6]. Tooth loss in adults has been associated with an impairment of the Oral Health-Related Quality of Life (OHRQoL), and the distribution of tooth loss affects the severity of the impairment [7]. Hence, preventive strategies to avoid tooth loss in elders is considered imperative.
It is essential to identify the carbohydrate-content in foods for the purpose of caries prevention [8]. Exposure of oral bacteria to fermentable carbohydrates leads to the production of organic acids [9,10]. Organic acids are known to decrease oral pH, initiating caries development at a critical pH value [9]. Research indicates that, compared to a carbohydrate-free diet [10], a carbohydrate-rich diet leads to a rapid pH decrease in plaque, which disturbs the equilibrium between the calcium and phosphate concentration in the enamel hydroxyapatite and plaque [9]. In-vitro-studies have demonstrated the initiation of demineralization of enamel was at pH < 5.5 [11 ], and of radicular dentine at pH < 6.7 [12].
Foodstuffs such as honey, sweets, beverages or drinking solutions, and pharmaceuticals have already been tested many years ago using the telemetry method, in order to prevent caries in the general population by means of education and product labeling (e.g. safe for teeth label) [13]. Even infant formulations have been identified to be highly cariogenic and potentially initiating early childhood caries [14]. However, industrially-manufactured ONS or senior formulas (Table 1) do not provide any information on their potential cariogenicity. Little or no evidence exists neither in experimental models nor in epidemiological studies on the cariogenicity of these products [15]. This is important, as dental caries, mainly root caries, is known to pose a major problem in geriatric patients [16]; and ONS are widely used in the treatment of malnutrition in this vulnerable group [17], especially in caredependent institutionalized elders with dementia [18].
The elderly population is increasing worldwide, and people aged over 80 years old are projected to increase to approximately 426 million by 2050 [3]. Along with this trend, an increase in the number of teeth retained to an advanced age has also been reported [4e6]. Although, retention of teeth in older adults is a positive trend, nevertheless it also augments the risk for caries or root caries and their associated complications [19,20]. Root caries is a common clinical finding in dependent older adults and is estimated that approximately 39% of institutionalized elders are affected [20,21]. Caries-related complications may include acute pain [22], which can be undiagnosed in elders with severe dementia. Furthermore, untreated deep caries lesions may lead to pulpal necrosis and abscesses, which may have serious adverse effects in dependent elders. Caries accounts for tooth loss which occurs later in life [23e26]; tooth loss coupled with malnutrition and other preexisting medical conditions can further affect the general healthand oral health-related Quality of life of the elders [27]. Moreover, clinical procedures to treat these conditions in dependent older adults are complex and are often associated with increased medical risks, administrative, as well as financial challenges [28]. Therefore, prevention of caries and root caries in older adults is cardinal in safeguarding oral health.
This study aimed to quantify the caries potential of randomly selected ONS products frequently administered in Switzerland and to compare it against a standard known cariogenic (10%-sucrose solution) [26]. The null hypothesis of this study was that the tested ONS products have no cariogenic potential.

Materials and methods
The study procedure received ethics approval (KEK-ZH-No. StV 02/14). The tested ONS are classified in Switzerland as Foods for Special Medical Purposes (FSMP) by the national Federal Office of Public Health (FOPH).
Ten ONS formulations were randomly selected and purchased from the local pharmacies as over-the-counter-products ( This in-vivo-study was performed under standard laboratory test conditions in one of the internationally recognized and calibrated test centers [13]. The degree of the potential cariogenic pH attack on enamel in pH x min was obtained using a validated telemetric measurement series of the pH of interdental plaque and fluid. These measurements were recorded with the ONS and a standard control (10% sucrose solution) [13]. An oral washout procedure was performed by requesting the participants to rinse with water in between the test and control measurements. Clinical relevance for tooth demineralization was set at pH < 5.7 (critical pH for enamel). Area-under-the-curve (AuC) for pH < 5.7 for the ONS AuC P pH<5.7 and for the control sucrose solution, AuC S pH<5.7 , were plotted. Relative cariogenicity (RC) potential was calculated using the formula RC ¼ AuC P pH<5.7 /AuC S pH<5. 7 .
The test assessed the hydrogen ion concentration (H þ ) under an undisturbed layer of plaque at the level of the enamel surface on the teeth. Habitual consumption or normal diffusion of substrates into the interdental areas and in plaque were not disturbed during the continuous pH measurements [29]. This method is a standard procedure applied for testing foods, beverages, snacks, and medicines [13]. The specification of the internationally standardized Table 1 Relative cariogenicities of the various oral nutritional supplement (ONS) products tested with 95%-confidence interval (95%-CI) and significance set at p < 0.05 (ANOVA) and RC influencing factors.

ONS Product
Flavor study protocol for "safe for teeth" products, recommends four repeated measurements with at least two participants to guarantee reliability of the measurements [13].
A cariogenic potential of the tested product is given when the maximal interdental pH drop is below 5.7 (critical pH for potential enamel damage in tests) after consumption.

Study protocol
Five volunteers (2 men and 3 women; mean age: 76.8 ± 9.15 years) were enrolled from the volunteer pool of the telemetry center in the Clinic of Conservative and Preventive Dentistry, Center of Dental Medicine at the University of Zurich in Switzerland. An informed consent was received from all participants. These participants were included if they satisfied the following inclusion criteria: missing teeth in the premolar and molar regions of the mandible, no active caries lesions, no untreated caries cavities, periodontal or other oral diseases, stimulated salivary secretion rate >0.6 ml/min [30] measured by the collection of whole saliva after chewing an inert gum base for 5 min, wearing a removable partial denture, no medications in medical records.
A test electrode in a partial denture, specially fabricated for each test subject, served to record the plaque-pH values measured for 30 min graphically and numerically. The measuring electrodes were installed interdentally distal to the canine tooth or the first or second premolar [31]. The tests were performed in a laboratory, which frequently performs such experiments for the industry in order to get the product certification "safe for teeth" [13]. The tests were sequenced and timed as described in a previously published similar experiment [29]. The participants were requested to refrain from toothbrushing and other oral hygiene procedures for at least 3e7 days before the testing, in order to allow sufficient biofilm (dental plaque) build up [13]. A reference electrode was placed on the right forearm. A measuring electrode was built in the respective partial dentures that were connected to a data recorder (Fig. 1). To neutralize the oral fluid the participants were requested to chew on paraffin wax for approximatively 2e3 minutes (Fig. 2, pc). The participants were then administered the ONS (Fig. 2, t) at room temperature and requested to maintain the solution in the mouth for 2 minutes. The telemetric pH measurements were then recorded for 30 minutes. A wash-out procedure was performed by serving the participants a rinse with water to neutralize the pH value (Fig. 2, w). The participants were requested to chew on paraffin again to produce neutral saliva. Following this the control solution (10%-sugar solution) was administered and the measuring procedures performed for the test solution were repeated to observe the functionality of the plaque (pH value must fall <5). The protocol allows one single test series (1 ONS and 1 control) per day. At the end of all tests, a dental hygiene procedure was carried out by a dental hygienist, which included prophylaxis along with a fluoride application on the participants' teeth.
Relative cariogenicity (RC) was expressed as a mean ± standard deviation. Data was verified for a normal distribution. ANOVA and post hoc tests (LSD test) were used to evaluate the difference in the RC between the products with the level of statistical significance set at p < 0.05. Ancillary analyses included the effects of the characteristic of ONS such as readily consumable (ready to use or mix and use), flavor (fruit or other), age of plaque (3e7 days), consistency (semi-solid or liquid), fiber containing (fiber-enriched or regular), protein containing (high protein or regular) and style (milk-based or fruit-based) on the RC.

Results
The mean RC of the various tested ONS with 95% confidence interval (95% CI) is shown in Table 1, along with the characteristics and categorizations of the products. The AuC P pH<5.7 ranged between 6.02 ± 4.34 and 30.57 ± 14.03 (Table 1). Statistical analysis revealed a significant difference between the various tested products AuC P pH<5.7 [F(9, 30) ¼ 3.183, p ¼ 0.008]. All ONS tested showed a potential cariogenicity on enamel. Relative cariogenicity was shown to be between 0. 31 Table 1 and are listed according to the decreasing RC potentials.

Discussion
As this study provides clear evidence that all the tested ONS are potentially cariogenic on enamel, the null hypothesis is therefore rejected. It is apparent that the results are most germane to conditions in the mouth of elderly or other risk groups (e.g. oncology). Interventional epidemiological studies would most probably confirm this. Administration of these ONS pose serious ethical concerns. Variability in the results were observed between the participants. This could be because of a few factors. Firstly, this variability could be because of the constitution of the individual himself. Different individuals demonstrate different caries risk susceptibility [32]. Therefore, the potential cariogenic risk imposed by each ONS can be varied. Then there is the fact that only 5 volunteers were tested and the measurements were repeated only four times for each participant. This low number of repeated observations might have contributed to an amplified result. However, it has been demonstrated by previous published studies that the number of participants used and the number of repeated observations performed in the current study is an acceptable protocol for such experiments [13]. Fig. 1. Telemetry volunteer with a specially manufactured removable partial denture in the mouth that is connected by a wire to the recording equipment. The influence from other parameters is secondary to the fact, that individual persons have a potential caries risk in addition to other known risk factors (e.g. low saliva buffer capacity). It is one of the pillars of caries pathogenesis that caries arises from the interaction between tooth (host), bacteria (plaque), carbohydrate (substrate) and the factor time [9]. Only a few studies have reported that ONS may have an impact on dental health [14,15,33]. It is important to appreciate that this test aimed to find even small effects and that there is an international consensus on the used telemetry procedures. Therefore, our telemetry results finding a cariogenic potential of the tested ONS indicate that they may have a meaningful in-vivoeffect. The results show a high cariogenic potential on enamel (pH < 5.7) and one must bear in mind that dentine is even more susceptible. This has been corroborated by an in-vitro-study by Castro and colleagues (2019) which demonstrated that a powdermilk ONS for seniors from a state-supported nutritional program in Chile, led to a high risk of developing root caries [15]. Basic dental research has demonstrated that the quantity of carbohydrate intake [34] is less related to caries development than the frequency [35]. Numerous literature in dentistry have emphasized the importance of limiting the frequency of sugar intake in between meals [8,34e38]. This, however, conflicts with the principles of clinical nutrition recommended by the clinical nutrition societies [1,39], and manufacturers [40]. Restriction of ONS to main meals [1]m a y not be an appropriate recommendation for a product, which is nutritionally essential to the prevention and treatment of malnutrition in elders and other risk groups. Results therefore strongly suggest that dental preventive measures during ONS therapy are indispensable to prevent caries development. Especially in dependent elders with dementia, dysphagia, and impaired mobility there is an urgent need for additional preventive measures. However, the compliance for oral hygiene in elders with dementia is a challenge per se. Given the rising trends for ONS prescription in market analyses [40e42], it is time to advance the preservation of dental health in parallel to the use of ONS. Moreover, it is essential that the ONS are administered based on individual needs to potentiate a maximum benefit and if possible to restrict the use of high carbohydrate containing ONS in the high-risk group for dental caries [44].
The results of this study demonstrate that factors like flavor, age of plaque, fiber content, consistency, style, and administration-type do not impact the RC of the tested products. The tested ONS show different RC potentials when compared to each other (Table 2). A more in-depth evaluation of the ONS was not possible because the Fig. 2. Representation of relevant areas of intraoral pH telemetry curves during a test series. P ¼ preparation of participant, pc ¼ paraffin chewing, t ¼ test product intake, w ¼ water rinsing, I ¼ 10% sucrose solution intake. product information on the package and on the product-homepage was lacking. Hence, precise influences from processed foods, carbohydrates and other nutrients were not identifiable [45]. Although the study did demonstrate that the RC potentials of high-protein containing ONS containing were lower, it could not be ascertained why this was so. It can only be speculated, at this juncture, that the high-protein containing ONS will be beneficial in providing a protective effect of proteins for prevention of caries as demonstrated by previous studies [45]. This however, needs to be confirmed by future studies. A recommendation of a specific product is not possible from our results. However, we can recommend that the RC potentials of the ONS must be considered before administering it to the elder. The ONS indicated for a particular clinical situation must be weighed against other possible options, and the product with the lowest RC within the same category must be prescribed (e.g. Resource 2.0 fibre: lowest cariogenicity among high-energy, fiber-enriched, high-protein products; Table 1). Actual research on preventive fluoride supplementation (5 ppm sodium fluoride) in higher carbohydrate-containing ONS shows promising in-vitro results [33], which still needs to be confirmed in-vivo. Hence, the possibility for the consumers and prescribers to choose a less cariogenic product [43], or respective changes of the industrial formulations must be recommended. Moreover, the findings of this study emphasize that the products should be labeled to state whether they are "safe for teeth" or not [46]. The telemetry method used in our study is a standard for labeling products according to the criteria of Toothfriendly International as "safe for teeth". There is no label for "not safe for teeth", but there should be an explicit call for oral health safeguarding measures in ONS consumers. In 2009, authors from a special interest group [47] have assumed that ONS are cariogenic. Our results support their assumption. The interest group recommends a multidisciplinary community nutrition team approach, an Oral Health Risk Assessment, patient information leaflets, a caregiver training on the impact of ONS on dental health, strategies to improve nutritional intake from food, and oral nutrition integrated care pathways.
The current study has several strengths, including a control, use of a validated objective, and an internationally accredited tool assessing potential cariogenicity on enamel, utilization of a largely accepted standardized scientific procedure, which yields sufficient power to detect effects in healthy subjects. Limitations include the fact that we did not test the effects on a larger vulnerable caries risk population. Finally, these results do not give an in-depth analysis of which carbohydrate in the formulation contributed most to the cariogenic potential.
ONS are usually given to the patients in between meals. This could be a contributing effect for the cariogenic potential as demonstrated by the Vipeholm Study (1954) [35]. The study demonstrated that the effects of repeated intake of cariogenic food in-between meals, in vulnerable subjects, is a high risk factor for the development of caries [35]. Therefore, along with the carbohydrate content, it is equitably imperative on the close monitoring of the frequency and the time of the administration of the ONS along with an efficient oral hygiene plan. Furthermore, manufacturers are encouraged to provide information about the carbohydrate content on the packaging of their respective ONS in order to aid the consumer as well as the clinicians in planning adequate preventive and hygiene measures.

Conclusions
The ONS administered to the elders and malnourished patients are potentially cariogenic. It is imperative that the administration of the ONS must be based on individual needs to potentiate a maximum benefit. Wherever possible, an attempt to limit the use of high-carbohydrate containing ONS must be practiced along with the adoption of suitable preventive measures to arrest the development and progression of dental caries.

Funding sources
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit-sectors.