Article Text

Original research
Association between animal protein intake, oral frailty and calf circumference in middle-aged and older adults: a cross-sectional analysis from the Shika study
  1. Fumihiko Suzuki1,2,
  2. Shigefumi Okamoto3,
  3. Shingo Nakai4,
  4. Sakae Miyagi5,
  5. Hiromasa Tsujiguchi2,4,6,
  6. Akinori Hara2,4,6,
  7. Thao Thi Thu Nguyen7,
  8. Yukari Shimizu8,
  9. Koichiro Hayashi2,
  10. Keita Suzuki6,
  11. Tomoko Kasahara6,
  12. Masaharu Nakamura2,
  13. Chie Takazawa2,
  14. Aya Ogawa2,
  15. Aki Shibata2,
  16. Takayuki Kannon9,
  17. Atsushi Tajima6,10,
  18. Hirohito Tsuboi11,
  19. Noriyoshi Ogino12,13,
  20. Tadashi Konoshita14,
  21. Toshinari Takamura15,
  22. Kuniko Sato16,
  23. Hiroyuki Nakamura2,4,6
  1. 1Department of Geriatric Dentistry, Ohu University School of Dentistry, Koriyama, Japan
  2. 2Department of Hygiene and Public Health, Faculty of Medicine, Institute of Medical, Pharmaceutical and Health Sciences, Kanazawa University, Kanazawa, Japan
  3. 3Laboratory of Medical Microbiology and Microbiome, Department of Clinical Laboratory and Biomedical Sciences, Division of Health Sciences, Osaka University, Suita, Osaka, Japan
  4. 4Department of Public Health, Graduate School of Advanced Preventive Medical Sciences, Kanazawa University, Kanazawa, Japan
  5. 5Innovative Clinical Research Center, Kanazawa University, Kanazawa, Japan
  6. 6Advanced Preventive Medical Sciences Research Center, Kanazawa University, Kanazawa, Japan
  7. 7Faculty of Public Health, Hai Phong University of Medicine and Pharmacy, Hai Phong, Viet Nam
  8. 8Faculty of Health Sciences, Department of Nursing, Komatsu University, Komatsu, Japan
  9. 9Department of Biomedical Data Science, School of Medicine, Fujita Health University, Toyoake, Japan
  10. 10Department of Bioinformatics and Genomics, Graduate School of Advanced Preventive Medical Sciences, Kanazawa University, Kanazawa, Japan
  11. 11Graduate School of Human Nursing, The University of Shiga Prefecture, Hikone, Japan
  12. 12Department of Environmental Medicine, Faculty of Medicine, Kochi University, Kochi, Japan
  13. 13Third Department of Internal Medicine, School of Medicine, University of Occupational and Environmental Health, Kitakyushu, Japan
  14. 14Division of Diabetes Endocrinology and Metabolism, Yachiyo Medical Center, Tokyo Women's Medical University, Shinjuku-ku, Japan
  15. 15Department of Endocrinology and Metabolism, Kanazawa University Graduate School of Medical Sciences, Kanazawa, Japan
  16. 16Laboratory of Clinical Cognitive Neuroscience, Graduate School of Medical Science, Kanazawa University, Kanazawa, Japan
  1. Correspondence to Dr Hiroyuki Nakamura; hiro-n{at}po.incl.ne.jp

Abstract

Objective To investigate the relationship between oral frailty (OF), nutrient intake and calf circumference (CC) in middle-aged and older adults.

Design Cross-sectional study.

Setting Residents of four model districts of Shika town, Ishikawa Prefecture, Japan, using data from November 2017 to February 2018.

Participants One hundred and ninety-four residents aged ≥50 years in four model districts of Shika town. The OF total score ≥3 was defined as OF. Participants were divided into OF and non-OF groups and divided into the low-CC/kg and the high-CC/kg groups.

Outcome measures The primary outcome is to use a two-way analysis of covariance to analyse the interaction between the two CC/kg groups and the two OF groups on nutrition intake. The secondary outcome is to use multiple regression analysis to investigate the nutrients significantly related to CC/kg when stratified by OF, with age, sex, body mass index, drinking status, smoking status and regular exercise as input covariates.

Results A two-way analysis of covariance revealed a significant interaction between the two CC/kg groups and the two OF groups on animal protein intake (p=0.039). Multiple comparisons using the Bonferroni analysis revealed a significantly lower animal protein intake in the OF group than in the non-OF group with a low CC/kg (p=0.033) but not in the group with a high CC/kg. The multiple regression analysis stratified by OF revealed a positive correlation between animal protein intake and CC/kg (p=0.002).

Conclusions The present results revealed a significantly lower animal protein intake in the OF group than in the non-OF group in the low-CC/kg group, but no such difference was observed in the high-CC/kg group. Further longitudinal studies are needed to elucidate this relationship.

  • Aging
  • EPIDEMIOLOGIC STUDIES
  • EPIDEMIOLOGY

Data availability statement

Data are available on reasonable request.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

STRENGTHS AND LIMITATIONS OF THIS STUDY

  • Calf circumference (CC/kg) was used as an index to consider the influence of obesity.

  • Epidemiological relationships among oral frailty (OF), CC/kg and nutrient intake were elucidated.

  • Due to the cross-sectional design, causal relationships cannot be elucidated.

  • OF assessment methods have no uniform standards, and its interpretation may vary from study to study.

Introduction

The relationship between oral frailty (OF)1 2 or poor oral health3–5 and systemic frailty is well known. According to a systematic review by Dibello et al2 assessing OF with 12 different indicators, items strongly associated with systemic frailty include few remaining teeth and chewing difficulties. Further, frailty is associated with sarcopenia, of which one of the indicators is calf circumference (CC).6 7 It has been reported that chewing8 9 and swallowing10 11 functional decline are associated with decreased CC. A cross-sectional study conducted on older adults has revealed that poor chewing function is significantly related to gastrocnemius muscle thickness, which is a component of the calf.8 Moreover, a cross-sectional study of middle-aged and older adults has indicated that CC was positively correlated with the cross-sectional area of the geniohyoid muscle as measured by ultrasound.11 However, the relationship between CC and OF,4 12 13 which comprehensively assesses oral functions, such as oral dryness and oral hygiene status in addition to chewing and swallowing, has not yet been studied in detail.

The relationship of the muscles with nutrients has been well documented.14–16 Specifically, protein and amino acids are substrates for muscle protein synthesis,14 calcium is involved in muscle contraction and relaxation,15 and vitamin D promotes muscle synthesis and calcium absorption.16 Both animal experiments in rats17 and randomised controlled studies in humans using dietary supplements18 have reported the association between the combination of protein, calcium, vitamin D and muscle mass. However, only a few studies have evaluated OF and nutrient intake,19 20 indicating the unclear association between nutrient intake and muscle with OF.

Since OF comprehensively assesses the decline in oral function, we hypothesised that a decrease in nutrient intake due to OF, especially nutrients involved in muscle synthesis, seems to be involved in the decreased CC than OF alone. Therefore, this study was conducted to epidemiologically elucidate the relationship between OF, nutrient intake and CC in middle-aged and older adults living in Shika town, Ishikawa Prefecture, Japan.

Methods

Study design, setting and participants

This study was conducted as part of the Shika study with residents of Shika town. Shika town had 20 845 residents, of whom 12 532 were aged ≥50 years as of February 2018.21 The Shika study is a community-based epidemiological study using data such as questionnaires and medical information.13 19 22 This study included 361 residents aged ≥50 years in four model districts (Tsuchida, Horimatsu, Togi and Higashi Masuho), having large populations in Shika town who agreed to participate. The questionnaire survey, CC measurement and number of teeth examinations were conducted on 361 individuals. The brief-type self-administered diet history questionnaire (BDHQ) was not participated by 166 participants, and 1 participant did not have energy records within 600–4000 kcal/day. Online supplemental figure 1 shows the participants’ selection criteria. We performed a statistical analysis on 194 individuals. The participants were not involved in the design, conduct or reporting of our research.

Data sources and variables

Health survey data were gathered from the Shika town residents from November 2017 to February 2018. Sex (1: female; 2: male), drinking status (1: non-drinking or drinking less than once a month; 2: drinking at least once a month), smoking status (1: non-smoker; 2: past smoker; 3: current smoker) and regular exercise (1: exercise at least 30 min/session and twice a week; 2: no) were investigated using a self-reported questionnaire. Age, body mass index (BMI) and CC were obtained from the Shika study’s medical check-up data. The CC was measured on both the right and left sides at the greatest dimension of the calf, and the average value of both sides was calculated.23 We chose CC/kg as an indicator, considering the influence of obesity.23

The OF evaluation followed our previous methodology.13 Briefly, chewing, swallowing and oral dryness domains of the Kihon Checklist,24 25 number of teeth, and brushing habits were evaluated. Several OF studies have used these domains of the KCL.12 26 27 Well-trained dentists assessed the number of teeth, excluding removable dentures, bridge pontics or dental implants. The evaluation points for each item were as follows: chewing domain (no: 0 points; yes: 2 points), swallowing domain (no: 0 points; yes: 2 points), oral dryness domain (no: 0 points; yes: 1 point), the number of teeth (≥20 teeth: 0 points; <20 teeth: 1 point) and brushing habit (at least twice a day brushing: 0 points; less than twice a day brushing: 1 point). The sum of each point was considered the OF total score. Additionally, a score of ≥3 was defined as OF.

Nutrient intake was assessed using the BDHQ.28 29 The BDHQ is a four-page structured questionnaire assessing the frequency of 58 foods and beverages consumed, commonly by the general Japanese population. The BDHQ estimates dietary intake in the last month using an ad hoc computer algorithm. Previous studies have demonstrated BDHQ validity in Japanese populations.28 29

Statistical methods

Participants were classified into the non-OF group (total scores of 0–2) and the OF group (total scores of ≥3). Participants were categorised into the low-CC/kg and the high-CC/kg groups. Missing values are eliminated by the selection criteria. Statistical software was IBM SPSS V.26 for Windows (IBM). Student’s t-test was used for the mean±SD variables and the χ2 test for the n (%) categories for the two groups comparison. A two-way analysis of covariance (ANCOVA) adjusted for age, BMI, drinking status and smoking status was performed to examine the main effects and interactions between the two CC/kg groups and two OF groups on nutrient intake. The two-way ANCOVA results were confirmed using multiple regression analysis, with CC/kg as the dependent variable, nutrient intake as the independent variable and stratified by OF. Variable selection was based on the forced input method. The significance level was set to 5%.

Sample size

We used the free software, G-power, to calculate the sample size. In the t-test for two independent groups, effect size, α error probability and power were set 0.5, 0.05 and 0.8, respectively. The total sample size and actual power were found to be 128 and 0.801. In the F-test for ANCOVA, effect size, alpha error probability, power, number of covariates and number of groups were set 0.25, 0.05, 0.8, 5 and 4, respectively. The total sample size and actual power were 129 and 0.801. For the F-tests for linear multiple regression regression, effect size, alpha error probability, power and number of predictors were set to 0.15, 0.05, 0.95 and 7, respectively. The total sample size and actual power were found to be 153 and 0.950, respectively. Therefore, the sample size of this study was confirmed to be sufficient.

Patient and public involvement

None.

Results

Participant characteristics

Table 1 shows the participant characteristics. Of the 194 participants, 102 were male and 92 were female. The participants’ mean (SD) age was 63.00 (6.67) years for males, which was not significantly different from that of 64.48 (8.57) years for females. BMI (p=0.001), CC (p<0.001), total energy (p<0.001), the proportion of drinking status (p<0.001) and smoking status (p<0.001) were significantly higher in males than in females. Alternatively, CC/kg (p<0.001), and intakes of protein (p<0.001), animal protein (p<0.001), vegetable protein (p<0.001), calcium (p<0.001), and vitamin D (p=0.042) were significantly higher among females. The OF total scores were 2.40 (1.67) and 1.98 (1.68) for males and females, respectively, with no significant difference in sex.

Table 1

Participants’ characteristics

Comparison of two of groups

Table 2 shows the comparison of the two OF groups. The mean age was significantly older in the OF group (66.46 (7.89) years) than that in the non-OF group (62.00 (6.99) years) (p<0.001). The proportion of chewing (p<0.001), swallowing (p<0.001) and oral dryness (p<0.001) domains, less than twice a day brushing (p=0.001), and the OF total score (p<0.001) were significantly higher in the OF group than in the non-OF group. The number of teeth (p=0.002) was significantly higher in the non-OF group than in the OF group.

Table 2

Comparison of the two OF groups

Comparison of two CC/kg groups

Table 3 shows the comparison of two CC/kg groups. The mean age was not significantly different between the low-CC/kg group (63.52 (7.24) years) and the high-CC/kg group (63.90 (8.11) years). BMI (p<0.001), CC (p<0.001), total energy (p=0.016), and the proportion of male sex (p<0.001), drinking status (p<0.001), smoking status (p<0.001), and less than twice a day brushing (p<0.001) were significantly higher in the low-CC/kg group than in the high-CC/kg group. Alternatively, CC/kg (p<0.001), and intake of protein (p=0.002), vegetable protein (p<0.001), and calcium (p<0.001) were significantly higher in the high-CC/kg group than in the low-CC/kg group.

Table 3

Comparison of the two CC/kg groups

Main effects and interactions between CC/kg and OF groups on nutrients intake

The low-CC/kg group was subdivided into 56 and 47 participants in the non-OF and OF groups, respectively. The high-CC/kg group was subdivided into 64 and 27 participants in the non-OF and OF groups, respectively (table 4). A two-way ANCOVA was used to examine the main effects and interactions between CC/kg and OF on nutrient intake after adjusting for age, sex, BMI, smoking and drinking status. Any of the nutrients on CC or OF has no main effect. Intake of protein (p=0.040) and animal protein (p=0.039) revealed a significant interaction between the two CC/kg groups and the two OF groups. The post hoc Bonferroni analysis was used for multiple comparisons, which revealed a significantly lower animal protein intake in the OF group than in the non-OF group in the low-CC/kg group (p=0.033), but not in the high-CC/kg group. Therefore, the animal protein intake is significantly lower with OF only when the CC/kg is lower.

Table 4

Two-way ANCOVA of the CC/kg and OF groups on nutrient intake

Relationship between CC/kg and animal protein intake stratified by OF

Table 5 shows the multiple regression analysis stratified by OF. Age (standardised partial regression coefficient (β): 1.025; p<0.001, 95% CI 0.008 to 0.011), male sex (β: −0. 274; p<0.001, 95% CI −0.150 to −0.055) and drinking status (β: 0.127; p=0.013, 95% CI 0.010 to 0.085) were significant independent variable in the non-OF group. Alternatively, age (β: 0.732; p<0.001, 95% CI 0.005 to 0.008) and animal protein intake (β: 0.148; p=0.002, 95% CI 0.004 to 0.015) were significant independent variables in the OF group. Therefore, animal protein intake was significantly positively correlated with CC/kg only with OF.

Table 5

Relationship between animal protein intake and CC/kg stratified by OF

Discussion

This study mainly revealed a significantly lower animal protein intake in the OF group than that of the non-OF group in the low-CC/kg group, but not between the OF and non-OF groups in the high-CC/kg group.

CC is considered an alternative anthropometric measurement30 by dual-energy X-ray absorptiometry (DEXA),31 32 CT33 34 and ultrasound.35 36 An autopsy-based evaluation by Tresignie et al37 reported that CC is a reliable predictor for muscle mass. A cross-sectional study by Santos et al38 suggested an equation to estimate low appendicular skeletal muscle mass as measured by DEXA from CC, gender, race and age. Alternatively, a review by Prado et al39 revealed that the CC measurement indicated that various confounding factors, such as age, BMI, ethnicity and oedema, may influence its measure and interpretation. Using direct CC measurements seems sufficient for studies of undernutrition and sarcopenia; however, the influence of obesity should be considered in epidemiological studies of community residents. Takamura et al23 evaluated insulin resistance and metabolic abnormalities in individuals with obesity by dividing CC (cm) by body weight (kg) as an adjusted value, thus we adopted and used the same method.

The univariate comparison of low CC/kg and high CC/kg in our results revealed no significant differences in the chewing and swallowing domains, which are components of OF. Conversely, it has been reported that chewing8 9 and swallowing10 11 functional decline are associated with decreased CC. One reason for the difference between our results and those of previous studies8–11 is that we adjusted CC by dividing it by body weight. Specifically, previous studies8–11 have focused only on the loss of muscle mass in the oral and CC from the perspective of sarcopenia and have not evaluated the relationship when the individuals are obese. Sarcopenic obesity should also be considered regarding loss of muscle mass.40 Therefore, the use of CC/kg as an indicator to investigate the relationship with OF contributed to the originality of our study. Another reason our results differed from previous studies8–10 is that our analysis was based on subjects aged 50 years and older, so the amount of muscle loss may have been less than in other studies of subjects aged 65 years and older.

The animal protein intake of OF was lower than that of the non-OF group only in the low CC/kg but with no difference in animal protein intake in the high-CC/kg group with and without OF. This difference in protein intake in the high-CC/kg group is presumably because the low-CC/kg group is easily affected by a lower protein intake. A randomised controlled trial involving males aged ≥70 years by Mitchell et al41 revealed a significant increase in lean body mass and leg strength with doubled protein intake. Additionally, a review by Nowson and O’Connell42 indicates that lower protein intake is associated with compensatory loss of muscle mass that occurs on lower protein intakes. The univariate analysis of our study revealed a significantly higher protein intake of the high-CC/kg group than that of the low-CC/kg group, supporting previous studies that have revealed a relationship between protein intake and muscle mass. Therefore, we speculate that the low-CC/kg and OF groups may have been related to decreased chewing and swallowing function due to reduced protein intake, especially animal protein, as a possible mechanism. Conversely, the high-CC/kg group seems unlikely to have reduced protein intake. OF includes not only chewing and swallowing function but also factors, such as oral dryness and oral hygiene,12 13 43 thus we speculate that OF in the high-CC/kg group was likely influenced by endpoints other than chewing and swallowing domains. Furthermore, animal protein was more related to CC/kg and OF than plant protein in our results, suggesting that the higher digestibility of animal protein was more likely to provide essential amino acids needed for muscle protein synthesis.44 These results indicate that active animal protein intake tends to be effective in preventing systemic flailing in OF with a decrease in CC/kg. The clinical implication is that OF prevention combined with nutritional instruction may play a role in maintaining walking function if it can prevent the decline in CC/kg. As a research outlook, we would like to examine through longitudinal studies whether the combined OF prevention and high animal protein intake can prevent the decline in CC/kg.

Limitations

Study limitations include its cross-sectional design, thus causal relationships cannot be elucidated. Additionally, OF assessment methods have no uniform standards, thus their interpretation may vary from study to study. Therefore, similar results might not be obtained if our data were analysed using other OF assessment methods. Additionally, we have not examined the reliability and validity of our method of assessing OF. Furthermore, the self-administered BDHQ may lack objectivity. Finally, we did not evaluate all items in BDHQ.

Conclusions

The results of this study revealed that the animal protein intake of the OF group was significantly lower than that of the non-OF group in the low-CC/kg group, but no such difference was observed in the high-CC/kg group. Further longitudinal studies are needed to elucidate this relationship.

Data availability statement

Data are available on reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by the Ethics Committee of Kanazawa University (No. 1491). Participants gave informed consent to participate in the study before taking part.

Acknowledgments

The authors thank all the participants in the Shika study and all of the field survey staff and experimental staff at the Department of Hygiene and Public Health, Faculty of Medicine, Institute of Medical, Pharmaceutical and Health Sciences, Kanazawa University.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Contributors The authors’ responsibilities were as follows: FS and HN: conceptualisation; TKannon and AT: Data curation; FS: Formal analysis; FS and HN: Funding acquisition; FS, SO, SN, SM, AH, TTTN, YS, KH, KSuzuki, TKasahara, MN, CT, AO, AS, HTsuboi, NO, TKonoshita, TT and KSato : Investigation; FS, TT and HN: Methodology; HTsujiguchi and HN: Project administration; HN: Resources; HTsujiguchi, AH and HN: Supervision; H Tsujiguchi and MN: Validation; FS: Visualisation; FS: Writing—original draft; SO, SN, SM, AH, TTTN, YS, KH, KSuzuki, TKasahara, MN, CT, AO, AS, TKannon and, AT, HTsuboi, NO, TKonoshita, TT, KSato and HN: Writing–review and editing. HN is the guarantor of the study. All authors read and approved the final manuscript.

  • Funding This work was supported by JSPS KAKENHI, grant numbers JP19H03882 and JP22K10533

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.