Exploring Oral Function, Protein Intake, and Risk of Sarcopenia: A Scoping Review

Abstract

Introduction:

Sarcopenia is loss of both muscle mass and function with age and is associated with inadequate protein intake. However, evidence to suggest an association with oral health is less clear.

Objective:

To scope peer-reviewed published evidence (2000–2022) pertaining to oral function in relation to sarcopenia and/or protein intake in older people.

Methods:

CINAHL, Embase, PubMed, and Scopus were searched. Included were peer-reviewed studies measuring oral function (e.g., tooth loss, salivary flow masticatory function, strength of muscles of mastication, and tongue pressure) and a measure of protein intake and/or a measure of sarcopenia (appendicular muscle mass and function). Full article screening was conducted by 1 reviewer with a random 10% screened in duplicate by a second reviewer. Relevant content pertaining to study type, country of origin, measures of exposure, and outcomes and key findings was mapped and the balance of data showing a positive versus null association of oral health with outcomes charted.

Results:

Of 376 studies identified, 126 were screened in full, yielding 32 included texts, of which 29 were original articles. Seven reported intake of protein and 22 reported measures of sarcopenia. Nine distinct oral health exposures were identified, with ≤4 studies relating to any one of these measures. Most data were cross-sectional in nature (27 studies) and from Japan (20 studies). The balance of data showed associations between tooth loss and measures of sarcopenia and protein intake. However, the balance of data pertaining to any association between chewing function, tongue pressure, or indices of oral hypofunction and sarcopenia was mixed.

Conclusion:

A broad range of oral health measures have been studied in relation to sarcopenia. The balance of data suggests that tooth loss is associated with risk, but data pertaining to the oral musculature and indices of oral hypofunction are mixed.

Knowledge Transfer Statement:

The findings of this research will increase awareness among clinicians of the amount and nature of evidence pertaining to the relationship between oral health and risk of compromised muscle mass and function, including data showing that loss of teeth is associated with increased risk of sarcopenia in older people. The findings highlight to researchers the gaps in the evidence and where further research and clarification of the relationship between oral health and risk of sarcopenia is warranted.

Introduction

An adequate dietary protein intake is important in maintaining muscle mass and function in older age (65+ y), thereby helping to mitigate sarcopenia (Paddon-Jones et al. 2015). Sarcopenia refers to the pathologic reduction in skeletal muscle mass and strength and, consequentially, loss of function (Muscaritoli et al. 2010; Cruz-Jentoft et al. 2014). It is diagnosed through having both low muscle strength and low muscle mass and is associated with adverse health outcomes, including falls and fractures, cardiac and respiratory diseases, reduced cognitive function, loss of independence, increased risk of hospitalization, and increased mortality (Senior et al. 2015). Sarcopenia affects 5% to 50% of older people depending on gender, age, pathological conditions, and diagnostic criteria (Papadopoulou 2020).

In older adults, reduced masticatory performance (e.g., tooth loss, reduced chewing and biting ability, low salivary flow, stomatitis, reduced tongue pressure) may affect the ability to consume adequate dietary protein (Sheiham et al. 2001; Mendonça et al. 2018). It can therefore be postulated that compromised oral function may contribute to nutritional deficits, including low intake of protein, that contribute to loss of muscle mass and the development of sarcopenia. However, sarcopenia is a whole-body disease and affects not only skeletal muscles but also the muscles involved in mastication and swallowing (Machida et al. 2017). Reduced tongue pressure is a symptom of sarcopenia (Buehring et al. 2013). However, tongue pressure is also reduced with age independently of sarcopenia (Machida et al. 2017; Sakai, Nakayama, Tohara, Kodama, et al. 2017), and age-related decrease in tongue pressure could affect dietary intake, thus contributing to risk of sarcopenia. Therefore, it can be postulated that the direction of effect between food intake and sarcopenia, as well as between some elements of oral health (e.g., tongue pressure, lip pressure, occlusal force and function, and strength of the muscles involved in mastication) and sarcopenia, may be bidirectional. With the overall aim of identifying what is known about the relationship between diet, oral function, and risk of sarcopenia, the objective was to conduct a scoping review to map existing data pertaining to this area and to identify potential gaps in the evidence. The specific aim was to scope the existing published evidence pertaining to compromised oral health, intake of dietary protein, and presence of sarcopenia. The specific objectives were to 1) to identify existing peer-reviewed evidence pertaining to any relationship between different elements of compromised oral health and i) risk of sarcopenia, ii) intake of dietary protein, and iii) both, and 2) to map available data by study type and volume of information, geographic location, measures of oral health and function, measures of sarcopenia employed, and measures of protein intake, along with an overview of findings and the balance of data showing an association versus no association. The outcomes will inform on existing evidence pertaining to oral function and risk of sarcopenia, whether there has been sufficient research in this area, and where gaps in data exist. The results will inform future development of research studies in oral health, protein intake, and sarcopenia.

Methods

The review was conducted according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) (Tricco et al. 2018) and Joanna Briggs Institute methodology for scoping reviews (Peters et al. 2021). The protocol was published on FigShare (Teo and Moynihan 2022).

Population, Context, and Concept

Participants were older adults (generally aged 65 y and over), including those diagnosed with type 2 diabetes mellitus or with under- or overnutrition. Excluded were studies with a specific focus on participants with acute or chronic diseases (e.g., cancer, kidney disease, liver disease).

The concept of interest was oral function in relation to sarcopenia and/or intake of dietary protein. Included were studies measuring oral function (e.g., by tooth loss, reduced salivary flow [dry mouth, xerostomia], masticatory function, bite force, and strength of muscles of mastication, tongue pressure) and a measure of protein intake (e.g., g/d, percent contribution to energy intake, achieving/not achieving intake reference values) and/or a measure of sarcopenia (which had to include a measure of both muscle mass and of muscle function). Studies measuring muscle function (i.e., handgrip strength, walking speed, physical frailty) without a measure of muscle mass were excluded. Excluded also were studies that measured dietary protein intake and risk of sarcopenia but without a measure of oral health/function. Studies that measured presbyphagia (sarcopenic dysphagia) without a measurement of general sarcopenia were excluded.

The context was data from populations in any country from any type of intervention or epidemiological study dated from 2000 to January 2022 to capture the global perspective published this century.

Types of Information

Included were peer-reviewed studies (randomized controlled trials [RCTs], nonrandomized trials, quasi-experimental studies, and reviews) that contained data relevant to the PCC with at least an abstract written in English. Protocols, abstracts, preprints, conference proceedings, thesis, gray literature, and any other non-peer-reviewed articles were excluded.

The PCC mnemonics (population/concept/context) was used to define all searches across the online databases. First, the concept was broken down into “oral function,” “sarcopenia,” and “dietary protein.” Alternative terminologies for these concepts were determined and recorded in a logic grid. This led to the generation of initial search terms. An initial scoping search was performed on PubMed using MeSH thesaurus terms and common terms to determine the search sensitivity and the need for more concept synonyms and variations to be added to the logic grid. This was replicated for the other databases (CINAHL, Embase, and Scopus) by searching for the major concepts and their synonyms, while identifying more concepts, synonyms, and variations to be added. The logic grids were adapted to work accordingly to the specific rules of each online database. The text words contained in the title and abstract of retrieved articles in the initial search were analyzed, in addition to all the MeSH/Emtree terms used to index the articles. Next, a second search using all identified keywords and indexed terms was conducted across all included databases. The final logic grids recorded for each online database are presented in the Appendix to the protocol (Teo and Moynihan 2022). The reference lists of included articles (after full-text screening) were searched for additional sources.

Study Selection

In terms of studies selection, PCC mnemonics (population/concept/context) guided the eligibility criteria for relevant studies. Resulting hits were imported into Covidence software (Veritas Health Innovation) after removal of duplicates. A primary screening of titles and abstracts was conducted by 2 independent researchers (JLT, PM) to eliminate articles that were clearly outside of the inclusion criteria. Any differences between the reviewers’ decisions were resolved through discussion. If consensus was not reached, a third-party reviewer was consulted.

A secondary screening on included full texts was conducted by 1 reviewer (JLT). A random 10% sample of all full-text studies was screened in duplicate by a second reviewer (PM), and interrater reliability was assessed. The reference list of included studies was also searched for additional studies. Reasons for exclusion were recorded at this stage. Any differences between reviewers’ decisions were resolved through discussion. If consensus was not reached, a third-party reviewer was consulted. The process of duplicate screening a 10% sample of articles was repeated until >95% agreement between reviewers was achieved. Articles excluded at full-text screening are presented in Appendix Table 1 along with reason for exclusion. The stages of article identification and screening are reported in a PRISMA-ScR chart (Fig. 1).

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) flowchart.

Data Charting/Mapping

Data charting of included studies was conducted by 1 reviewer and checked by a second reviewer. During data charting, a mapping spreadsheet was developed using Microsoft Excel to ensure a systematic data-mapping process. The key information mapped included author(s), year of publication, origin/country of origin (where the source was published or conducted), aims/purpose, population and sample size, exposure variables and how these were measured (i.e., oral health measures), relevant outcome variables and how these were measured (i.e., methods to assess risk of sarcopenia and dietary protein intake), duration of the study (for intervention and cohort studies), and key findings that relate to the topic (PCC). These data are given in Appendix Table 2. Charting of the articles was trialed by 1 reviewer for an approximately 10% sample of identified studies and was modified accordingly to include relevant data across all articles. The results were described narratively in tabulated form, with the balance of studies showing positive or null associations between the oral health exposure and the outcome (sarcopenia/low protein intake) presented in harvest plots (Fig. 2).

Figure 2.

Harvest plot to illustrate the balance of studies showing an association between compromised oral health indices and (A) protein intake (denoted by dark bars) and (B) indicators of sarcopenia (denoted by lighter bars).

Results

Figure 1 presents the PRISMA-ScR flowchart. Of 376 studies identified, 126 were included for full-text screening, of which 93 were excluded (see Appendix Table 1), resulting in 32 included texts. This included 3 narrative reviews/opinion pieces (Azzolino et al. 2019; Fujishima et al. 2019; Watanabe et al. 2020) that did not include any original data. Remaining studies (n = 29) were grouped according to review questions into those that reported on the impact of oral health on 1) intake of protein (n = 7) (Sheiham et al. 2001; Zhu and Hollis 2014; Okada et al. 2015; Iwasaki et al. 2016; Mendonça et al. 2018; Nagano et al. 2020; Ram et al. 2020) and 2) measures of sarcopenia (n = 22) (Okada et al. 2010; Moriya et al. 2011; Gaszynska et al. 2014; Murakami et al. 2015; Gaszynska et al. 2017; Iwasaki et al. 2017; Machida et al. 2017; Sakai, Nakayama, Tohara, Kodama, et al. 2017; Sakai, Nakayama, Tohara, Maeda, et al. 2017; Shiraishi et al. 2017; Takahashi et al. 2018; van den Heuven et al. 2019; Yamaguchi et al. 2019; Abe et al. 2021; Chang et al. 2021; Kugimiya et al. 2021; Matsuo et al. 2021; Nishioka et al. 2021; Shimizu et al. 2021; Yoshida et al. 2022; González-Fernández et al. 2021). No studies were identified that reported the association between oral health measures and both intake of protein and a measure of sarcopenia. Of the 7 studies that reported on the impact of oral health on intake of protein, 1 was an intervention study (Nagano et al. 2020), 1 was a cohort study (Iwasaki et al. 2016), and 5 studies had cross-sectional data (Sheiham et al. 2001; Zhu and Hollis 2014; Okada et al. 2015; Mendonça et al. 2018; Ram et al. 2020). Three studies were of Japanese populations (Okada et al. 2015; Iwasaki et al. 2016; Nagano et al. 2020), 2 of UK populations (Sheiham et al. 2001; Mendonça et al. 2018), 1 of a US population (Zhu and Hollis 2014), and 1 from New Zealand (Ram et al. 2020). All 22 studies that reported on an association between oral health exposure and measures of sarcopenia included cross-sectional analysis (see Appendix Table 2). The majority of these data came from studies of Japanese older adults (n = 17) (Okada et al. 2010; Moriya et al. 2011; Murakami et al. 2015; Iwasaki et al. 2017; Machida et al. 2017; Sakai, Nakayama, Tohara, Kodama, et al. 2017; Sakai, Nakayama, Tohara, Maeda, et al. 2017; Shiraishi et al. 2017; Takahashi et al. 2018; Yamaguchi et al. 2019; Abe et al. 2021; Kugimiya et al. 2021; Matsuo et al. 2021; Nishioka et al. 2021; Shimizu et al. 2021; Yoshida et al. 2022). Data also came from Poland (n = 2) (Gaszynska et al. 2014; Gaszynska et al. 2017), Spain (n = 1) (González-Fernández et al. 2022), Taiwan (n = 1) (Chang et al. 2021), and the United Kingdom (n = 1) (van den Heuven et al. 2019). There were no data from the Americas, Africa, or Australasia. Table 1 summarizes information on the various oral health exposures that have been explored in relation to protein intake or measures of sarcopenia (muscle mass and muscle strength) and the frequency with which these measures were applied across studies. Table 2 groups studies by outcome (protein intake/sarcopenia) and summarizes each study, including country, aims, population, oral health exposure, outcome measures, and key findings.

Table 1.

Summary of Measures of Exposure (Oral Health) and Outcome (Protein Intake and/or Measure of Sarcopenia) with the Number of Studies Reporting the Application of the Measure.

Variable Reported	Number of Studies
Oral health exposure measure
Tongue pressure	6
Lip strength	1
Masticatory muscle function/thickness	4
Occlusal force	2
Number of teeth (including Eichner’s index)	10
Comparison of wearing dentures vs. being dentate	3
Denture condition/fit	1
Chewing ability (objective measure)	3
Chewing ability (subjective/perceived)	3
Oral hypofunction (criteria of the Japanese Society of Gerodontology)	3
Revised Oral Assessment Guide (ROAG)	2
Oral Health Assessment Tool (OHAT), Japanese version	1
Measure of sarcopenia applied
HGS and SMI using bioelectrical impedance	10
Measure of protein intake	7
Presence or absence of sarcopenia using criteria of the Asian Working Group on Sarcopenia criteria   (no other detail specified)	4
HGS and body cell mass index	2
HGS and upper arm muscle area	1
HGS and calf circumference	6
HGS, calf circumference, and SMI	1
SARC-F questionnaire	1

HGS, handgrip strength; SARC-F, Strength, assistance with walking, rising from a chair, climbing stairs, and falls questionnaire. SMI, Skeletal Muscle Mass Index.

Table 2.

Summary of Included Studies.

Aims/Population	Oral Health Measure (Exposure)	Outcome Measures Relevant to the Concept of This Review	Key Findings Relevant to This Review
Studies relating to compromised oral health and protein intake
Intervention studies
Nagano et al. 2020 (Japan)
Aim: investigate the impact of physical intervention and nutritional intake on tongue strength and swallowing function. n = 95, orthopedic rehabilitation patients aged >65 y with sarcopenia.	Tongue pressure	Protein intake and EI measured during a physical rehabilitation intervention.	Protein >1.2 g/kg/d and EI >30 kcal/kg/d associated with increased tongue pressure.
Cohort studies
Iwasaki et al. 2016 (Japan)
Aim: to investigate whether impaired dentition was associated with declined nutrient intake. n = 286 community-dwelling adults aged 75 y followed up 5 y.	FTU: pairs of opposing natural/prosthetic teeth: ≤5 units = “impaired” dentition.	Dietary intake assessed using validated FFQ at baseline and at 5-y follow-up. Intake protein.	21.3% had impaired dentition (n = 65). Those with impaired dentition had a greater decline in protein intake (P < 0.05).
Cross-sectional studies/data
Sheiham et al. 2001 (United Kingdom)
Aim: to assess the relationship between dental status and nutrient intake. Representative sample n = 418 dentate and 287 edentulous. Independently living adults aged 65+ y.	Distribution and number of natural teeth, grouped as 0–10, 10–20, and 21+. Dentate compared with edentulous groups.	4-d weighed food intake to determine intake of protein (and other nutrients) in g/d.	Dentate had a higher mean intake of protein (68 g/d) compared with edentulous (60.1 g/d): regression coefficient, 3.4 (95% CI, 0.9–5.8; P = 0.007). Protein intake: 0–10 teeth: 65 g 10–20 teeth: 68 g 21+ teeth: 70 g (regression coefficient, 0.33; 95% CI, 0.06–0.60; P = 0.02).
Zhu and Hollis 2014 (United States)
Aim: to investigate association between the number of natural teeth and nutrient intake. N = 914 adults aged 19+ y (NHANES survey).	Number of teeth from NHANES oral exam data set. Participants classified into: (1) Full dentition (28 teeth) (2) Moderate dentition (21–27 teeth) (3) Poor dentition (20 teeth or fewer)	1 × 24-h dietary recall. Protein intake.	Protein intake positively associated with the total number of natural teeth (P < 0.05). Adjusting for SES, physical activity, smoking status, and EI.
Okada et al. 2015 (Japan)
Aim: to investigate the association between lower-extremity motor function and occlusion and whether protein intake mediates the association. Septuagenarians (n = 655); octogenarians (n = 629).	Bilateral maximal occlusal force.	Protein intake by validated, self-administered brief diet history questionnaire.	Protein intake was associated with occlusal force (standardized direct effect, 0.10; P = 0.001) and occlusal force was associated with walking speed (standardized direct effect, 0.11; P = 0.001).
Mendonça et al. 2018 (United Kingdom)
Aim: to investigate the prevalence and determinants of low protein intake. Community-dwelling older people (n = 722) aged 85 y in 2006).	Tooth count (self-reported).	Protein intake (g/d and % energy) assessed by 2 × 24-h recalls. Low protein intake defined as <0.8 g/kg/d. Adequate intake ≥0.8 g/kg/d.	Tooth count predicted higher protein intake (β = 0.003; SD, 0.001; P = 0.001). Swallowing problems predicted low protein intake (β = 0.04; SD, 0.023; P = 0.077). Participants with low protein vs. adequate had a lower tooth count (5.3 [SD, 8.0] vs. (6.7 [SD, 8.5]; P = 0.013) (unadjusted analysis).
Ram et al. 2020 (New Zealand)
Aim: to investigate intake and sources of protein n = 214 Māori (aged 80–90 y), n = 360 non-Māori (aged 85 y) in LiLACS cohort.	Wear dentures (Y/N), biting and chewing problems, swallowing problems (self-reported).	2 × 24-h recall assessed protein intake dichotomized as adequate using the NZ-NRV and EAR (>0.75 g/kg/d for women and >0.86 g/kg/d for men).	Dentate or partial dentures were more likely to have an adequate protein than full dentures. No denture compared with complete (OR, 1.335; 95% CI, 0.763–2,336); partial vs. complete denture (OR, 1.948; 95% CI, 1.170–3,244), P = 0.036. Swallowing/chewing problems NS.
Studies relating to compromised oral health and measures of sarcopenia
Cross-sectional studies/data
Okada et al. 2010 (Japan)
Aim: to examine relationships between chewing ability and anthropometric measurements or nutritional status. N = 200 older adults (78 men and 122 women), mean age 76.6 ± 17.1 y, from geriatric clinical settings.	Masticatory performance measured via color-changeable chewing gum.	Upper AMA and HGS.	Correlations between HGS and AMA and chewing ability were 0.25 (P = 0.001) and 0.11 (P = 0.15), respectively. High chewing ability group had higher AMA, 34.5 ± 10 cm2 vs. 29.7 ±10.1) cm2, P = 0.002; and HGS, –21.8 ± 7.7 kg vs. 16.6 ± 7.8 kg, P = 0.001.
Moriya et al. 2011 (Japan)
Aim: to determine if self-assessed masticatory ability is independently related to HGS and SMM. N = 381 adults aged 67–74 y.	Self-perceived chewing ability: Good = “I can chew all foods” Fair = “I can chew slightly hard foods” Poor = “I can only chew soft and pureed foods”	HGS; SMM using bioelectrical impedance.	Perceived chewing ability associated with HGS (independently of SMM) but not SMM. HGS was as follows: Good: 27.9 ± 7.1 kg Fair: 27.0 ± 6.9 kg Poor: 24.0 ± 6.2 kg, P = 0.012 SMM was as follows: Good: 21.6 ± 4.4 kg Fair: 21.4 ± 4.5 kg Poor: 21.3 ± 4.4 kg, P = 0.845.
Gaszynska et al. 2014 (Poland)
Aim: to investigate association between MMT, dental status, and physical fitness parameters. n = 259 (97 men, 162 women) care home residents; mean age, 75.3 ± 8.9 y.	Self-perceived chewing ability, number of teeth, number of opposing pairs of teeth, number of number of posterior tooth pairs. MMT by ultrasonography.	HGS. Timed up and go test. BCMI (body cell mass/height m2) using bioelectrical impedance.	Positive correlations between the number of teeth and BCMI (r = 0.23, P = 0.017) and HGS (r = 0.16, P = 0.01). HGS: weak MMT, 16.4 ± 6.0; strong MMT, 24.2 ± 8.0. BMCI: weak MMT, 6.8 ± 0.9; strong MMT, 8.1 ± 1.3; P < 0.001.
Murakami et al. 2015 (Japan)
Aim: to investigate the relationship between chewing ability and sarcopenia in addition to known sarcopenia-related factors. n = 761 community-dwelling adults (mean age, 73.0 ± 5.1 y).	Chewing ability (color-changing gum), number of teeth, number of functional teeth, occlusal force.	SMM and HGS, usual walking speed used to classify participants into presarcopenic or sarcopenic groups (based on guidelines of the European Working Group on Sarcopenia in Older People using the cutoff values of the Asian Working Group for Sarcopenia).	Number of teeth, occlusal force, and chewing ability were lower in sarcopenia. Number of teeth: 17.5 ± 9.4 vs. 20.3 ± 8.8 Occlusal force: 407 ± 280 N vs. 551 ± 347 N Chewing ability: 69.8% good vs. 88.7% good (P < 0.05).
Gaszynska et al. 2017 (Poland)
Aim: to evaluate the effect of age and age-related factors (e.g., dentition, muscle strength, and nutrition) on masticatory muscles, electromyographic activity during chewing. N = 30 older women aged 68–92 y.	Dentition: natural dentition vs. dentures (complete and partial).	BCMI (using bioelectrical impedance). HGS (handheld dynamometer).	Complete denture wearers had a lower HGS and BCMI compared with partial dentures or natural teeth. HGS (kg): Complete: 13.7 ± 3.9 Partial: 13.2 ± 5.1 Natural: 14.8 ± 4.8 BMCI (kg/m2): Complete: 6.1 ± 0.8 Partial: 6.3 ± 0.9 Natural: 6.5 ± 1.2 MMT: 3.7 mm—all groups.
Iwasaki et al. 2017 (Japan)
Aim: to test the association of dental status with sarcopenia. n = 272 community-dwelling adults aged ≥75 y.	Number of natural teeth and occluding pairs of natural teeth (dental exam). Denture status evaluated in those with partial dentures.	Sarcopenia using criteria of the Asian Working Group on Sarcopenia (Chen et al. 2014): HGS, <26 kg men and <18 kg women; low gait speed (<0.8 ms−1); and low SMI (<7.0 kg ms−2 men; <5.7 kg ms−2 women).	No occluding pairs of natural teeth at higher risk of sarcopenia compared with having ≥10 occluding pairs. Adjusted OR, 3.37 (95% CI, 1.07–10.61). Ill-fitting dentures (compared with well fitting) had higher risk of having sarcopenia (adjusted OR, 5.07; 95% CI, 1.59–16.19).
Machida et al. 2017 (Japan)
Aim: to examine how aging and sarcopenia affect tongue pressure and jaw-opening force. n = 97 men and 100 women aged 78.5 ± 6.6 and 77.8 ± 6.2 y, respectively.	Tongue pressure.	Presence of sarcopenia (using the criteria of the Asian Working Group on Sarcopenia), SMM (using bioelectrical impedance), HGS, and gait speed.	Tongue pressure lower with sarcopenia: In men, 21.1 ± 4.2 kg vs. 29.1 ± 7.8 kg, P < 0.001. In women, 21.0 ± 5.2 kg vs. 26.5 ± 7.4 kg, P = 0.001. Sarcopenia (and age) was an independent determinant of tongue pressure: Men: β-coefficient, –5.47 (95% CI, –8.25 to –2.60), P = 0.001 Women: β-coefficient, –3.80 (95% CI, –7.16 to –0.79), P = 0.015.
Sakai, Nakayama, Tohara, Kodama, et al. 2017 (Japan)
Aim: to investigate whether tongue strength is associated with muscle function, nutritional status, and dysphagia. n = 64 men, 110 women; median age, 84 (IQR, 80–89) y with sarcopenia, in rehabilitation hospital.	MTP.	HGS, CC, and AMA. Sarcopenia-related factors: 1) Activity evaluated using the Barthel index 2) MNA-SF, BMI, and controlling nutritional status (CONUT) 3) Disease, evaluated by the presence of cachexia and CRP levels.	Significant correlation between MTP and muscle mass (CC, AMA), HGS, and activity. MTP independently associated with HGS (and MNA-SF), after adjusting for age. MTP not correlated with muscle mass in multiple regression analysis.
Sakai, Nakayama, Tohara, Maeda, et al. 2017 (Japan)
Aim: to clarify the relationship between tongue strength, lip strength, and NRS. n = 201 adults aged >65 y admitted to a rehabilitation hospital.	Lip strength and tongue strength.	NRS defined by the presence of both malnutrition (MNA-SF) and sarcopenia diagnosed via HGS and SMM.	Tongue strength (median (IQR)) in NRS = 22.9 kPa (17.7, 27.7) vs. normal = 29.7 kPa (24.8, 35.1). Lip strength in NRS = 7.7 N (5.6, 9.8) vs. normal = 9.9 N (8.4, 12.3). Multivariate logistic regression showed NRS associated with tongue strength (OR, 0.93; 95% CI, 0.87–0.98) and lip strength (OR, 0.76; 95% CI, 0.66–0.88), P < 0.001.
Shiraishi et al. 2017 (Japan)
Aim: to investigate the association between poor oral status with rehabilitation outcome. n = 108 patients, mean age 80.5 ± 6.8 y (50.9% men) admitted to convalescent rehabilitation wards.	ROAG score to classify with/without oral problems: Score <9: without Score 9+: with	HGS and CC.	HGS with oral problems: 15.1 (range, 0–21.5) Kg. HGS without oral problems: 19.8 (range, 15.1–21.9) kg, P = 0.037. CC with oral problems: 29.6 ± 3.0) cm. CC without oral problems: 32.0 ± 3.3) cm, P = 0.026.
Takahashi et al. 2018 (Japan)
Aim: to clarify the prevalence of sarcopenia in dental clinic outpatients and its relationship with oral health status. n = 279 (173 female) adults, mean age 76 ± 7.5) y.	OHAT (assessment tool based on 8 factors: lips, tongue, gum, tissues, saliva, natural teeth, cleanliness, pain) (Chalmers et al. 2005).	Participants with sarcopenia compared with those without. Sarcopenia diagnosed using criteria of the Asian Working Group for Sarcopenia Criteria.	OHAT score higher (poor) with sarcopenia: 6.0 (3–8) vs. 3(2–4), P < 0.001.
van den Heuven et al. 2019 (United Kingdom)
Aim: to investigate the association between food neophobia, physical disadvantage, and demographic characteristics. n = 377 adults aged 55+ y.	Self-reported (questionnaire) use of dentures scored: No dentures = 1 Partial denture = 2 Complete denture = 3	Participants dichotomized into risk of sarcopenia/no risk using SARC-F questionnaire (Malmstrom and Morley 2013).	Denture wearing and risk of sarcopenia (assessed by SARC-F) were significantly correlated (no further details provided).
Yamaguchi et al. 2019 (Japan)
Aim: to examine the relationship of MMEI with skeletal muscle, physical function, and nutrition status, to determine whether MMEI is an indicator of these parameters. n = 139 (65 men and 74 women) community-dwelling older people.	Eichner’s index to classify tooth loss status. MMEI and Masseter muscle thickness.	SMI (bioimpedance) and CC for skeletal muscle mass and HGS.	HGS independently related to MMEI, CC and SMI not related to MMEI (multiple regression analysis with MMEI as dependent variable).
Abe et al. 2021 (Japan)
Aim: to examine number of teeth and masticatory function as factors in the pathogenesis of sarcopenia and diabetes mellitus. n = 635 community-dwelling adults aged 40–74 y from rural areas.	Number of teeth. Masticatory function (chew gummy bear 15 s and count number of pieces).	HGS and SMI (bioelectrical impedance). Possible sarcopenia status (based on Asian Working Group for Sarcopenia; Chen et al. 2020).	Tooth number and masticatory function were negatively associated with low HGS (and possible sarcopenia status, based on function only) but not associated with SMI.
Chang et al. 2021 (Taiwan)
Aim: to explore how tongue pressure level relates to malnutrition. n = 326 community-dwelling adults aged >65 y.	Maximal tongue pressure using the Iowa Oral Performance Instrument (IOPI; Northwest Co., LLC).	Body composition using an InBody 270 multifrequency body composition analyzer (Biospace), SMI, and HGS.	No significant correlation between tongue pressure and SMI (r = 0.08, P = 0.1) or HGS (r = 0.10, P = 0.06).
Kugimiya et al. 2021 (Japan)
Aim: to clarify the relationship between oral hypofunction and sarcopenia. n = 268 men, n = 610 women, mean age 76.5 ± 8.3 y.	Oral hypofunction defined as 3 of 7 of the following: oral hygiene, oral dryness, occlusal force, tongue-lip, motor function, tongue pressure, masticatory function and swallowing function (as defined by JSG; Minakuchi et al. 2018).	HGS and SMI used to classify with/without sarcopenia.	Oral hypofunction was associated with sarcopenia. Sarcopenia occurred at an increased frequency in participants with oral hypofunction: OR, 1.59 (95% CI, 1.02–2.47).
Matsuo et al. 2021 (Japan)
Aim: to explore the association of oral hypofunction with physical characteristics and function. To test the effects of a program including oral and physical exercises and textured lunch gatherings on oral and physical function. n = 86 community-dwelling adults aged 65+ y with oral hypofunction.	Classified as with or without oral hypofunction based on criteria of the Japanese Society of Gerodontology 2016 (Minakuchi et al. 2016).	SMI (bioimpedance) and HGS. Sarcopenia defined using criteria of the Asian Working Group for Sarcopenia (Chen et al. 2014). Participants randomly assigned to a 12-wk comprehensive oral, physical, and eating intervention or a control (physical activity only).	Sarcopenia prevalence did not differ between oral hypofunction (8% ± 15.1%) and normal (1% ± 3.0%), P = 0.144. SMI: Oral hypofunction: 6.0 ± 0.7 kg/m2 Normal: 6.2 ± 0.8 kg/m2 HGS: Oral hypofunction: 20.5 ± 4.3 Normal: 23.4 ± 5.0, P = 0.06
Nishioka et al. 2021 (Japan)
Aim: to clarify prevalence of Co-MS and its associated factors. n = 601 adults aged 65+ y on rehabilitation wards.	ROAG to evaluate oral functions. Components include voice, swallowing, lips, teeth/dentures, mucosa, gingiva, tongue, and saliva.	Risk of malnutrition screened using the MUST and BMI. Sarcopenia assessed using HGS and SMI. Patients were classified as having malnutrition, sarcopenia, or both (Co-MS).	ROAG scores associated with higher odds for Co-MS in the crude analysis but did not remain a significant factor in adjusted analysis. OR, 0.01 (95% CI, 0.92–1.11), NS.
Shimizu et al. 2021 (Japan)
Aim: to investigate if patients admitted to a rehabilitation hospital, with possible/probable sarcopenic dysphagia, have different clinical characteristics. n = 129 adults aged 65+ y.	Tongue pressure <20 kPa = probably sarcopenic dysphagia Tongue pressure 20+ kPa = possible sarcopenic dysphagia).	SMI and HGS.	Low tongue pressure had lower HGS. HGS: Men with low tongue pressure: 13.7 ± 5.2 kg Men without low tongue pressure: 18.6 ± 5.5 kg, P = 0.04 Women with low tongue pressure: 9.9 ± 4.6 kg Women without low tongue pressure: 11.2 ± 4.4 kg, P = 0.046 SMI: Men with low tongue pressure: 5.66 ± 0.78 kg/m2 Men without low tongue pressure: 6.15 ± 0.83 kg/m2, P = 0.24 Women with low tongue pressure: 4.40 ± 0.73 kg/m2 Women without low tongue pressure: 4.64 ± 0.72 kg/m2, P = 0.15.
Yoshida et al. 2022 (Japan)
Aim: to examine the frequency of oral hypofunction in older people and to examine any relationship with frailty and sarcopenia. n = 340 community-dwelling people aged 75 y.	Participants dichotomized into those with and without oral hypofunction as defined by JSG.	Presence or absence of sarcopenia based on the Asian Working Group on Sarcopenia Criteria. SMI (bioelectrical impedance). HGS.	Prevalence of sarcopenia was 2.6% (n = 9) in those with oral hypofunction vs. 7.6% (n = 23) in those without. P = 0.012.
González-Fernández et al. 2022 (Spain)
Aim: to explore the relationship between sarcopenia, malnutrition, and dependence with MMT. n = 464 adults aged >65 y in nursing homes for 6+ mo.	MMT	HGS, appendicular skeletal muscle mass (bioimpedance) to confirm. Physical performance measured by gait speed for severe sarcopenia. MNA questionnaire.	Each 1-mm decrease in MMT increased the risk of sarcopenia by 57% (OR, 0.43), the risk of malnutrition (by MNA) by 63% (OR, 0.37) adjusted for confounding variables (age, sex, Barthel index, and diet texture).

AMA, arm muscle area; BCMI, body cell mass index; BMI, body mass index; CC, calf circumference; Co-MS, coexistence of malnutrition and sarcopenia; CI, confidence interval; CRP, C-reactive protein; EAR, estimated average requirement; EI, energy intake; FFQ, food frequency questionnaire; FTU, functional tooth units; HGS, handgrip strength; IQR, interquartile range; JSG, Japanese Society of Gerodontology; LiLACS, Life and Living in Advanced Age Cohort Study; MMEI, masseter muscle echo intensity; MMT, masticator muscle tension; MNA, Mini Nutritional Assessment; MNA-SF, Mini Nutrition Assessment Short Form; MTP, maximum tongue pressure; MUST, Malnutrition Universal Screening Tool; NHANES, Nutrition and Health Examination Survey; NRS, nutrition-related sarcopenia; NS, not significant; NZ-NRV, New Zealand Nutrition Reference Values; OHAT, Oral Health Assessment Tool; OR, odds ratio; ROAG, Revised Oral Assessment Guide; SARC-F, strength, assistance with walking, rising from a chair, climbing stairs, and falls questionnaire; SES, socioeconomic status; SMI, Skeletal Muscle Mass Index; SMM, skeletal muscle mass; Y/N, yes/no.

For each measure of compromised oral function, the balance of studies indicating a positive versus null association between this measure and risk of sarcopenia is presented in Figure 2.

Discussion

The aim of this study was to scope the peer-reviewed literature pertaining to the relationship between compromised oral health and risk of sarcopenia and/or intake of dietary protein and to map available data by study type and volume of information, geographic location, measures of oral health and function, measures of sarcopenia employed, and measures of protein intake, along with an overview of findings and the balance of data showing an association versus no association. Overall, the findings show that despite the identification of a substantial body of evidence overall (contained in 29 articles), the evidence included a diverse number of measures of oral health that related either to the status of the dentition or to the oral musculature. No more than 4 studies related to any 1 oral health exposure and an outcome (sarcopenia or protein intake). Most data pertaining to the relationship between measures of compromised oral health and sarcopenia are cross-sectional in nature and from Japan, with limited data from the United States and Europe and no data from Latin America, Australia, or Africa, despite age-related sarcopenia being a global condition.

The balance of data shows an association between sarcopenia with loss of natural teeth, wearing dentures, lower occlusal force, and masticatory muscle thickness. The balance of data also shows that the number of natural teeth is associated with protein intake. The data suggest no association between presence of sarcopenia and perception of reduced chewing function or reduced masticator muscle function, with data on objective measures of chewing function showing mixed findings (Fig. 2). The balance of data pertaining to studies measuring tongue pressure or indices of oral hypofunction (e.g., Oral Health Assessment Tool [OHAT], Revised Oral Assessment Guide [ROAG], and oral hypofunction as define by Japanese Society of Gerodontology (JSG) was also mixed. A concept diagram summarizing balance of data pertaining to the association between oral health and sarcopenia is presented in Figure 3.

Figure 3.

Concept diagram summarizing the balance of data pertaining to the association between oral health measures and sarcopenia.

The findings relating to number of teeth, which included longitudinal data, suggest that tooth loss may have an adverse effect on the intake of dietary protein; this could contribute to loss of muscle mass and function. This suggests that preserving natural dentition has a role to play in the prevention of sarcopenia through enabling the consumption of adequate dietary protein. Research has shown that intervention with dietary protein coupled with physical activity can increase muscle protein synthesis. To maintain muscle protein synthesis, a protein intake of 1.2 g/kg/d with 20 to 25 g with each meal has been suggested (Bollwein et al. 2013; Cruz-Jentoft et al. 2014), yet data identified in this review showed a lower tooth count was associated with protein intakes below 0.8 g/kg/d (Mendonça et al. 2018), and wearing dentures was associated with not meeting the estimated average requirement (the amount estimated to be adequate for 50% of a population) of 0.75 g/kg body weight (Ram et al. 2020). More attention is needed to support patients with tooth loss to consume adequate protein intake. Tooth loss may also affect other dietary factors such as intake of fiber and of fruits and vegetables (a good source of antioxidants), which were not explored in this review but probably are dietary risk factors in the etiology of sarcopenia (Welch et al. 2020). The findings cannot, however, rule out that those experiencing tooth loss eat a less healthy diet throughout life, which could place them at increased risk of developing sarcopenia in later life, independently of oral health status.

Only 3 studies included objective measures of chewing function, with 2 studies showing a positive association (Okada et al. 2010; Murakami et al. 2015) and 1 no association (Abe et al. 2001). These mixed findings may relate to the measure of chewing function used: Abe et al. (2001) counted bolus particles following chewing a gummy bear, whereas the other studies used color-changing chewing gum, which may give a more precise measure.

Over a quarter of studies reporting measures of sarcopenia assessed oral status using mixed-variable indices, such as the OHAT (Chalmers et al. 2005), ROAG (Ribeiro et al. 2014), and oral hypofunction (as defined by the JSG; Minakuchi et al. 2018). The mixed balance of data from these studies (3/6 reporting an association) may reflect the diverse measures included in such assessments. For example, in the assessment of oral “hypofunction,” having 3 of 7 criteria (relating to oral hygiene, oral dryness, occlusal force, tongue-lip, motor function, tongue pressure, masticatory function and swallowing function) classifies a person as having oral hypofunction. It is possible, therefore, that the nature of oral hypofunction in some people relates predominantly to dental-related elements (oral hygiene, masticatory function dryness), whereas for others, it relates predominantly to oral muscle function. Further exploration of data from these studies could help elucidate the relative impacts of tooth loss versus muscular impact, which would help to determine the direction of effect. The mixed balance of data relating to reduce tongue pressure and sarcopenia was not anticipated as it has been postulated that sarcopenia is a whole-body disease affecting the muscles of the oral cavity as well as skeletal muscles (Shimizu et al. 2021), and therefore one would expect sarcopenia to proportionally affect the muscles of mastication and skeletal muscles. Further scrutiny of data, including quality assessment (which was outside the remit of this scoping review), may help to shed light on the mixed findings. Moreover, longitudinal data are needed to elucidate the direction of effect between loss of tongue pressure and skeletal sarcopenia in order to establish whether oral health intervention to maintain tongue pressure has a role to play in the prevention of sarcopenia.

Authoritative definitions of sarcopenia, defined by the European Working Group on Sarcopenia in Older People (Cruz-Jentoft et al. 2014) and the Asian Working Group on Sarcopenia (Chen et al. 2020), stipulate that diagnosis of sarcopenia should include a measure of low muscle mass and either a measure of low muscle strength or low physical function, and both provide culturally appropriate cutoff values. Therefore, to be included in the current review, studies required a measure of muscle quantity and strength or function regardless of the methods used to assess these parameters. Most included studies used HGS to assess muscle function; however, a variety of methods assessed muscle mass (Table 1). The measures used may have affected the accuracy and precision; for example, muscle mass assessment measured using bioelectrical impedance is likely to be a more precise measure compared with anthropometric measures such as limb circumference. The use of different measures makes cross-study comparisons complicated; however, the aim of this review was to map the volume, nature, and direction of effect and not to assess effect size. Meta-analysis of data in any future systematic review would require uniformity in the definition of sarcopenia and its measurement.

Limitations of the Scoping Review

This scoping review only included data published since 2000 and therefore may have missed evidence published before this cutoff date. The review also did not include the gray literature and non-peer-reviewed articles such as preprints and abstracts. Moreover, the limited resources available for this research dictated that only articles that had an English abstract were included and did not allow for translation of full articles not published in the English language, meaning that some data may have been missed.

Future Research

Much of the existing data are cross-sectional in nature, and more longitudinal data and intervention studies are required to clarify whether the association between oral hypofunction and sarcopenia is a cause of effect or bidirectional. Elucidating the direction of effect will inform on whether oral intervention has a role to play in the prevention of sarcopenia. Moreover, because no studies were identified that reported the association between oral health measures and both intake of protein and a measure of sarcopenia, future studies should consider exploring if intake of protein and/or other dietary factors mediate any association between tooth loss and risk of sarcopenia.

Despite being included as a characteristic of oral hypofunction, this review identified no study specifically investigating any association between oral dryness and risk of sarcopenia. Oral dryness affects masticatory performance (Pedersen et al. 2002) and therefore is likely to affect ability to eat and nutritional intake. This warrants further investigation. This review specifically explored data pertaining to the condition of sarcopenia, and therefore articles reporting on the symptoms of physical frailty only but without reference to a measure of both muscle function and muscle mass were excluded: 47 articles fell into this category (Appendix Table 1).

Most of the available data pertaining to the association between oral health and sarcopenia identified in this review come from Japan (>75% of studies). However, with a growth in the aged population, sarcopenia is an increasing global public health issue: a recent systematic review estimates the global prevalence at approximately 10% in populations aged 60+ y (Shafiee et al. 2017). However, there is some suggestion that prevalence is considerable higher (~20%) in non-Asian populations (Shafiee et al. 2017). Therefore, analysis of data from cohort studies of older populations from a diversity of countries is needed to fully understand the significance of oral health in aging well from a global perspective.

Conclusion

This study has identified and mapped the broad range of oral health measures that have been studied in relation to the condition of sarcopenia. The balance of data shows that tooth loss is associated with risk, but data pertaining to oral musculature are mixed. More research, including data from longitudinal and intervention studies, especially from geographical locations outside of Japan, is needed to fully elucidate if oral health intervention has a role to play in preventing the condition of sarcopenia.

Author Contributions

P.J. Moynihan, contributed to the conception and design, data interpretation of the data, drafted the manuscript; J-L. Teo, designed the search strategy and mapping charts, data interpretation of the data, critically revised the manuscript. All authors gave final approval and agree to be accountable for all aspects of the work.