The Journal of the American Dental Association
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Am Dent Assoc, Vol 139, No 1, 63-71.
© 2008 American Dental Association
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nishimura, M.
Right arrow Articles by Shimono, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nishimura, M.
Right arrow Articles by Shimono, T.

RESEARCH

JADA Continuing Education

Using a Caries Activity Test to Predict Caries Risk in Early Childhood

Michiko Nishimura, DDS, PhD, Takashi Oda, DDS, Naoyuki Kariya, DDS, PhD, Seishi Matsumura, DDS, PhD and Tsutomu Shimono, DDS, PhD


   ABSTRACT
TOP
 ABSTRACT
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Background. The authors conducted a two-year longitudinal study to show the predictive abilities of a caries activity test (Cariostat, Dentsply-Sankin, Tokyo), and to include the predicted screening indexes that were based on previous caries activity test results and lifestyle factors that influence caries activity.

Methods. The subjects were 1,206 children born in 2000. These children participated in health examinations at 18 months, 2 years and 31/2years of age at Kurashiki-City Public Health Center in Kurashiki-City, Japan. Two of the authors performed caries activity tests at 18-month and 2-year examinations. Questionnaires regarding the patient’s lifestyle were mailed to each participant’s parents or guardians. The authors analyzed these questionnaires to evaluate lifestyle factors that made participants susceptible to caries.

Results. A caries activity test score at 18 months of age not only reflected caries incidence but also predicted caries incidence and screening results in 2- and 31/2-year-old children. A caries activity test score at 2 years of age both reflected and predicted children’s caries incidence and screening results at 31/2years of age. Breast-feeding and use of the bottle to intake liquids other than water produced significant caries susceptibility in 18-month-old children. Additionally, increased frequency and total time of sucrose intake put 2-year-old children at high risk of developing caries and failure of parental brushing produced a high risk in 31/2-year-old children.

Conclusions. A caries activity test could predict 31/2-year-old children’s caries risk based on 18-month and 2-year-old test results. Early weaning, less sucrose intake and toothbrushing by parents were effective in reducing a child’s caries risk.

Clinical Implications. The caries activity test is more useful than oral examination because it can indicate the need for caries-preventive treatment before a carious lesion actually is manifest.

Key Words: Dental caries; dental caries activity tests; dental caries susceptibility; incidence

Abbreviations: dft: Decayed and filled teeth. • dt: Decayed teeth. • NPV: Negative predictive value. • PPV: Positive predictive value. • ROC: Receiver operating characteristic. • SP: Specificity. • SRCC: Spearman rank correlation coefficient. • ST: Sensitivity.

Because caries is preventable, the diagnosis of caries as a tooth lesion is not sufficient for a treatment plan. The clinician must devise a need-related treatment plan on the basis of certain scientific diagnosis. In doing so, he or she should assess a patient’s caries risk status, because it provides an estimate of future caries activity.1 Therefore, an accurate caries risk assessment in early childhood is a necessary prerequisite to effectively formulating a "total health" strategy.2

It is difficult to identify caries-susceptible children on the basis of a visual and tactile oral examination. Many bacteriological caries activity tests have been developed to avert this difficulty. They are classified into two types: count methods3–5 and evaluation of bacterial virulence.6–8 Borgström and colleagues9 reported that the most common method used to identify caries-susceptible people is estimating the number of cariogenic bacteria such as lactobacilli and mutans streptococci in saliva or plaque samples taken from the patient. However, the power of bacterial counts to explain and predict a person’s risk of developing caries has not been sufficient. Borgström and colleagues’ recommendation was to evaluate one virulence factor, such as acidogenicity of these bacteria, in attempting to identify a caries-susceptible person.

The Cariostat (Dentsply-Sankin, Tokyo) caries activity test is a colorimetric test developed by Shimono and Sobue in 1974.10 The test medium contains sucrose and two kinds of pH indicators to display the continuous pH decrease of the test medium caused by micro-organisms gathered in the patient’s plaque sample. The microorganisms in the dental plaque metabolize sucrose and produce acids that react to these pH indicators, thus leading to colorimetric change. Some researchers have reported strong correlations between pH and the caries activity test score.11–13 Nishimura and colleagues14 reported positive correlations between caries activity test score and the counts of mutans streptococci and lactobacilli. Dental caries is well-known as having a number of causes.1 Therefore, a correct diagnosis depends on the acquisition of additional data including information on caries activity, lifestyle and oral hygiene habits.

A correct diagnosis of caries depends on the acquisition of data including information on caries activity, lifestyle and oral hygiene habits.

We determined to conduct a two-year longitudinal study to show the caries-reflective and -predictive abilities of the Cariostat test, and to include the predicted screening indexes based on previous caries activity test results and lifestyle factors that influence caries activity. Some researchers have already reported the caries activity test’s predictive value.11,15,16 However, there are no studies to show current and predicted screening indexes for each score of caries activity tests. A predictive and precise caries risk assessment is absolutely necessary to enable the clinician to evaluate a patient’s caries risk and develop a treatment plan that focuses on prevention.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Subjects. The subjects were 1,208 children born in 2000. These children underwent health examinations at 18 months, 2 years and 31/2 years of age at Kurashiki-City Public Health Center, Kurashiki-City, Japan. We excluded two of the 1,208 children because they did not receive the caries activity test at 18 months of age. We had already excluded from this study children taking any kind of medication during any of their two caries activity test sampling periods.

Methods. Oral examination. The 18-month, 2-year and 31/2-year-old children’s examinations included an oral examination, a caries activity test and counseling by Kurashiki-City public dental hygienists that was based both on responses to questionnaires regarding patients’ lifestyles and on previous caries activity test results. Each examiner was blinded as to previous caries activity test results. Initial caries (white-spot lesions) was not counted as a caries lesion. The caries activity test was not performed at the 31/2-year-old children’s oral examination.

The health examinations of 18-month- and 31/2-year-old children are routine and are sanctioned by the Japanese government, while the examinations of 2-year-old children are additional ones carried out at Kurashiki-City. The routine government-sanctioned examinations began in 197717; Kurashiki-City’s 2-year oral examinations began in 1988. Each examiner assessed dental caries according to the criteria of the Japanese Ministry of Health and Welfare’s Health Policy Bureau, so this was an integrated study of different systems. The same dentists, members of the Dental Society of Kurashiki-City, carried out the examinations in 18-month- and 31/2-year-old children; the 2-year-old children were examined by pediatric dentists (M.N. and T.O.) of Okayama University (Japan). Calibration of the oral examinations was impossible because of the different systems. However, the authors nevertheless considered this study to be important in predicting individual caries risk on the basis of current caries activity test results. The same pediatric dentists gathered plaque samples for the caries activity test from the 18-month and 2-year-old children. There was no institutional review board at Okayama University at the time the study was conducted; the examinations were performed under the auspices of the government and Kurashiki-City’s health program. Furthermore, Kurashiki-City completely sealed subjects’ private information and then released the data to us in coded form.

The caries activity test. The examiners collected plaque samples from the maxillary buccal cervical surfaces using sterile cotton swabs. The examiners ran the swab along the tooth surface five times in a swiping motion before placing it in an ampule containing 2 milliters of the Cariostat test medium. Dental hygienists incubated this medium at 37°C for 48 hours. They assigned a caries activity test score with reference to four standard colors. The same hygienists evaluated each of these scores as follows: score 0 (pH 5.8–7.2), 1.0 (pH 5.4 ± 0.3), 2.0 (pH 4.8 ± 0.3) and 3.0 (pH < 4.4). In this study, we used a modified scale in which the intervals between 0–1.0, 1.0–2.0 and 2.0–3.0 were divided into halves. The same dental hygienist evaluated all caries activity test results. We sent the test results to the subjects by mail.

Questionnaires. We administered questionnaires to the subjects’ parents or guardians to evaluate which factors regarding the patients’ lifestyles made them susceptible to caries. We mailed these questionnaires to subjects along with a notification of their examination date. The subject’s parents or guardians completed the questionnaire and brought it to the Kurashiki-City Public Health Center on the day of the child’s scheduled examination. The questions were as follows:

– Do you check and brush your child’s teeth?
– How many times a day does your child ingest sucrose-containing foods?
– Do you determine the total time of your child’s sucrose-containing food intake?
– Does your child continue to breast-feed or drink liquids other than water through a bottle?

Predicted screening indexes. The 18-month caries activity test results of children aged 2 and 31/2 years were predictive of the children’s either being caries-free or experiencing caries. We calculated predicted screening indexes on the basis of 18-month-old caries activity test results by using discriminate analysis for screening indexes. We calculated 31/2-year-old children’s predicted screening indexes on the basis of 2-year-olds’ caries activity test results in a similar manner.

Statistical methods. We analyzed the data released by the Kurashiki-City Public Health Center. We used Spearman rank correlation to evaluate the relationship between the caries activity test of the subject’s oral condition at 18 months and 2 years of age, as well as the predictive ability of the caries activity test. We used the Wilcoxon signed rank test to analyze the differences between caries activity test result distributions at 18 months and 2 years. Furthermore, we used the Mann-Whitney U test to analyze the caries activity test result distributions between caries-experienced and caries-free groups at both ages. We used discriminate analysis to create screening indexes for the results of the caries activity test. Screening indexes include sensitivity (ST), specificity (SP), validity (ST plus SP), positive predictive value (PPV) and negative predictive value (NPV). ST is the probability that a true high-risk child is predicted to be at high risk. SP is the probability that a true low-risk child is predicted to be at low risk. PPV is the probability that a child is truly at high risk when he or she is predicted to be at high risk. NPV is the probability that a child is truly at low risk when he or she is predicted to be at low risk. We used the {chi}2 test to investigate which factors regarding the subject’s lifestyle made him or her susceptible to caries. We considered a P value of less than .05 to be significant.

There were positive correlations between caries activity test results and mean number of carious teeth at 18 months and 2 years of age.


   RESULTS
TOP
 ABSTRACT
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
There were no significant differences in the distributions between caries activity test results when the children were 18 months old and 2 years old (Figure 1Go). Table 1Go shows the screening indexes of 18-month-old children and the 2- and 31/2-year-old children’s predicted indexes based on the 18-month-old caries activity test results. Table 1Go also shows the screening indexes of 2- and 31/2-year-old children’s predicted indexes based on the 2-year-old caries activity test results. Validity beyond 1.0 was accepted and used for the screening test. The validity of caries activity test score 0.5 was only 1.0, so we did not use this score as a cutoff point for dividing children into different caries-risk groups. Receiver operating characteristic (ROC) curves showed the results of screening indexes except for PPV and NPV. None of the curves for ST and SP was drawn in the meaningful area. (The meaningful area is beyond a diagonal line and the meaningless area is below a diagonal line. A diagonal line shows the validity of 1.0, so this line is included in the meaningless area.) The 2-year-olds’ current and predicted curves based on the 18-month-old caries activity test results were not the same. However, both curves were located beyond the diagonal line. Both predicted curves for the 31/2 year-olds based on their caries activity test results at the other ages were drawn close to each other (Figure 2Go).


Figure 1
View larger version (37K):
[in this window]
[in a new window]

  		Figure 1. The differences between the distributions with caries activity test results at the 18-month and 2-year time points (the Wilcoxon signed rank test was used).

 

View this table:
[in this window]
[in a new window]

  		TABLE 1 Screening indexes of each caries activity test score at every age, based on test results for 18-month-old and 2-year-old children.

 

Figure 2
View larger version (49K):
[in this window]
[in a new window]

  		Figure 2. Receiver operating characteristic (ROC) curves of 18-month-old and 2-year-old children and predicted curves based on the caries activity test results at both ages.

 
There were positive correlations between caries activity test results and mean number of carious teeth at 18 months and 2 years of age. The 2-year caries activity test results showed a stronger correlation with the mean number of decayed teeth than did the 18-month results (Table 2Go). None of the children had filled teeth at either of those ages. Table 2Go shows the caries prevalence at each caries activity test point. Caries prevalence was higher at 18 months than at any other caries activity test point. The 2-year-olds’ caries prevalence increased according to the caries activity test results. Table 3Go shows the mean number of 2- and 31/2-year-olds’ decayed and filled teeth (dft) at every caries activity test point in comparison with previous test results. Previous caries activity test results significantly predicted future caries incidence. The 18-month-old caries activity test results predicted 2- and 31/2-year caries incidences more strongly than the 2-year caries activity test results predicted the 31/2-year test results.


View this table:
[in this window]
[in a new window]

  		TABLE 2 Relationship between caries activity test results and actual mean number of carious teeth and caries prevalence at each test.

 

View this table:
[in this window]
[in a new window]

  		TABLE 3 Relationship between caries activity test results and predicted mean number of decayed and filled teeth and caries prevalence.

 
Table 3Go also shows the future caries prevalence based on previous caries activity test results. In 2-year-old children, the actual and predicted caries prevalences were similar from caries activity test score 0.5 to scores 2.0 and 3.0, but the actual caries prevalence was higher than the predicted one at caries activity test scores 0 and 2.5. The 31/2-year-olds’ predicted caries prevalence based on the 18-month and 2-year caries activity test scores were well-matched except for the 18-month caries activity test score of 3.0. A child who had a caries activity test score of 3.0 when he or she was 18 months old would have a higher probability of carious teeth at 31/2 years of age than would a 2-year-old child who had a 3.0 caries activity test score.

We analyzed the responses to the lifestyle questions for behaviors that would put children at risk of developing caries. We observed a significant difference between the children who reportedly continued breast-feeding or received anything other than water through a bottle through the age of 18 months. Additionally, we observed a significant difference regarding the frequency of sucrose intake and total time of sucrose intake through 2 years of age. There were no data regarding the act of parents’ brushing of the children’s teeth at 2 years of age. However, there was a significantly high rate of caries in children whose parents brushed the children’s teeth and who had a high sucrose intake at 31/2 years of age (Table 4Go).


View this table:
[in this window]
[in a new window]

  		TABLE 4 The distribution of caries activity test results and answers to questions regarding oral habits and lifestyle of caries-experienced and caries-free children.*

 

   DISCUSSION
TOP
 ABSTRACT
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Other researchers have investigated the reflective and predictive abilities of the caries activity test. Matsumura and colleagues17 reported that the caries activity test could screen high caries activity in children aged 1 year to 15 years, and they found a significant correlation between the caries activity test scores and number of decayed teeth for each age. Sutadi and colleagues11 and Tsubouchi and colleagues15 reported that the caries activity test had high screening indexes through their longitudinal studies in groups with high caries prevalence (more than 70 percent). Koroluk and colleagues16 reported the screening indexes of caries activity test, and their results were almost the same as ours. They used the same screening test but produced different screening indexes. The caries prevalence of the subjects in the study by Koroluk and colleagues was 33.3 percent; in our study, it was 32.4 percent. The subjects’ caries prevalence influenced the screening indexes.

The greatest difference between our study and other studies is that ours shows the screening indexes regarding each score. It is important to establish the screening indexes of each score for clinicians and researchers because they then can fix a cutoff point for patients on the basis of the shown screening indexes. High ST and low SP mean that there is a high probability of true positives and false negatives. Low ST and high SP means that there is a high probability of false positives and true negatives. Therefore, the practitioners can determine a cutoff point for their patients under the following circumstances: if patients’ caries prevalence is high, ST is more than SP; if patients’ caries prevalence is low, SP is more than ST. However, the test score that has the highest validity generally is considered to be the cutoff point.

Caries is a complex chronic disease,18 and the constantly alternating process of demineralization and remineralization19 on the tooth surfaces is the most clinically important process. Therefore, a caries activity test is considered better in screening for caries potential than for caries experience. It is important that the clinician screen for caries potential, especially before finding caries on the tooth. Therefore, the test validity beyond 1.0 is a prerequisite condition, because the basic goal in dentistry and medicine is to prevent the initiation of disease and its further development. In this study, a caries activity test score at 18 months of age not only reflected the 18-month-old caries status but also predicted the child’s caries status at 2 and 31/2 years of age. We saw the same results with the 2-year-old caries activity test scores. Furthermore, 2- and 31/2-year-old screening indexes were predicted on the basis of the 18-month-old caries activity test scores, and 31/2-year-old screening indexes were predicted on the basis of the 2-year-old caries activity test scores. The 31/2-year-olds’ predicted screening indexes based on the 18-month-olds’ and 2-year-olds’ caries activity test scores were similar because the two predicted ROC curves were similar. This means that there were no significant differences in the distributions of the 18-month-olds’, 2-year-olds’ and 31/2-year-olds’ caries activity test scores.

ROC curves generally are used to compare screening tests targeted at the same disease. An ROC curve is drawn by connecting the coordinates on an ROC curve plane with a normal curve. However, in this study we used a straight line instead of a normal curve to clarify the relationship of the curves’ ups and downs. Researchers and clinicians must adopt a cutoff point that is based on exact scientific evidence. To our knowledge, this study is the first study to show the screening indexes at each caries activity test point. A researcher or clinician can use the caries activity test to diagnose caries susceptibility on the basis of exact scientific evidence. An ultimate decrease in the prevalence of caries makes it imperative that the 18-month-old caries activity test result be improved. Investigation of 31/2-year-olds’ lifestyles may be useful for caries prevention in the future. Oral examinations of six-month-old infants, especially involving instruction of parents regarding appropriate oral hygiene practices, is considered to be effective in caries prevention. Tooth eruption and change of feeding method (from breast or bottle to weaning) occur in many children at this age. Pediatric dentists and dental hygienists can give parents or guardians information regarding children’s oral hygiene, diet and lifestyle factors that lead to higher caries rates. Parental awareness can prevent children from developing caries.20–22 Intake of fermentable carbohydrates, particularly sucrose, is well-known as a caries risk factor. The American Academy of Pediatric Dentistry’s recommendations include parental oral hygiene such as parents’ brushing of children’s teeth,21 and some researchers have reported that intake of liquid other than water is not acceptable.20,22,23 Therefore, the questions in this study were important, and the information concerning early childhood caries found in the results of our study can be effective in improving 18-month-olds’ caries activity test results.


   CONCLUSIONS
TOP
 ABSTRACT
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
The caries activity test succeeded in predicting 31/2-year-old children’s caries risk assessment based on 18-month and 2-year test results. Early weaning, less sucrose intake and toothbrushing by parents lowered a child’s caries risk. It also may be useful to show screening indexes of not only each score of the caries activity test but also of other caries activity tests so that clinicians and researchers can determine a cutoff point.


   FOOTNOTES
 

Dr. Nishimura is an assistant professor, Behavioral Pediatric Dentistry, Okayama University, 2-5-1 Shikata-cho, Okayama City, Okayama, 700-8525, Japan, e-mail "naruto10@md. okayama-u.ac.jp". Address reprint requests to Dr. Nishimura.


Dr. Oda is a clinical fellow, Behavioral Pediatric Dentistry, Okayama University, Japan.


Dr. Kariya is an assistant professor, Behavioral Pediatric Dentistry, Okayama University, Japan.


Dr. Matsumura is an associate professor, Behavioral Pediatric Dentistry, Okayama University, Okayama, Japan.


Dr. Shimono is a professor, Behavioral Pediatric Dentistry, Okayama University, Japan.


The authors are greatly indebted to Joel H. Berg, MS, for his kind suggestions and sophisticated English instruction.


   REFERENCES
TOP
 ABSTRACT
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 

  1. Reich E, Lussi A, Newbrun E. Caries-risk assessment. Int Dent J 1999;49(1):15–26.[Medline]

  2. Koch G. Importance of early determination of caries risk. Int Dent J 1988;38(4):203–10.[Medline]

  3. Larmas M. A new dip-slide method for counting of salivary lactobacilli. Proc Finn Dent Soc 1975;71(2):31–5.[Medline]

  4. Matsukubo T, Ohta K, Maki Y, Takeuchi M, Takazoe I. A semi-quantitative determination of Streptococcus mutans using its adherent ability in a selective medium. Caries Res 1981;15(1):40–5.[Medline]

  5. Jordan HV, Laraway R, Snirch R, Marmel M. A simplified diagnostic system for cultural detection and enumeration of Streptococcus mutans. J Dent Res 1987;66(1):57–61.[Abstract/Free Full Text]

  6. Snyder ML. A simple colorimetric method for the diagnosis of caries activity. JADA 1941;28:44–9.

  7. Alban A. An improved Snyder test. J Dent Res 1970;49(3):641.[Free Full Text]

  8. Maki Y, Yamamoto H, Matsukubo T, Takazoe I, Sibuya M, Asama K. Prevalence and caries activity test scored Resazurin disk method. J Dent Health 1984;34(2):18–26.

  9. Borgström MK, Sullivan A, Granath L, Nilsson G. On the pH-lowering potential of lactobacilli and mutans streptococci from dental plaque related to the prevalence of caries. Community Dent Oral Epidemiol 1997;25(2):165–9.[Medline]

  10. Shimono T, Sobue S. A new colorimetric method for caries diagnosis. Dent Outlook 1974;43(6):829–35.

  11. Sutadi H, Huey JC, Nishimura M, Matsumura S, Shimono T. The determination of the predictive value of caries activity test and its suitability for mass screening in Indonesia. Pediatr Dent J 1992;2(1):73–81.

  12. Huey JC, Nishimura M, Matsumura S, Shimono T. Comparison of mutans streptococci count methods and Cariostat test for caries risk assessment. Pediatr Dent J 1995;5(1):31–42.

  13. Rodivic OD. A longitudinal study of approximal caries in primary molars: predictive value of Cariostat. Pediatr Dent J 1996;6(1):125–34.

  14. Nishimura M, Bhuiyan MM, Matsumura S, Shimono T. Assessment of the caries activity test (Cariostat) based on the infection levels of mutans streptococci and lactobacilli in 2- to 13-year-old children’s dental plaque. ASDCJ Dent Child 1998;65(4):248–51, 229.

  15. Tsubouchi J, Yamamoto S, Shimono T, Domoto PK. A longitudinal assessment of predictive value of a caries activity test in young children. ASDC J Dent Child 1995;62(1):34–7.[Medline]

  16. Koroluk L, Hoover JN, Komiyama K. The sensitivity and specificity of a colorimetric microbiological caries activity test (Cariostat) in preschool children. Pediatr Dent 1994;16(4):276–81.[Medline]

  17. Matsumura S, Shimono T, Morisaki I, Shimono T, Sobue S. Dental caries experience and dental caries activity by new caries susceptibility test (CARIOSTAT) by the children in Okinawa prefecture (Tarama Island of Miyako). Jpn J Ped Dent 1980;18:612–7.

  18. Hunter PB. Risk factors in dental caries. Int Dent J 1988;38(4):211–7.[Medline]

  19. Silverstone LM. Remineralization phenomena. Caries Res 1977;11(supplement 1):59–84.[Medline]

  20. Johnsen DC, Gerstenmaier JH, Schwartz E, Michal BC, Parrish S. Background comparisons of pre-31/2-year-old children with nursing caries in four practice settings. Pediatr Dent 1984;6(1):50–4.[Medline]

  21. Febres G, Echeverri EA, Keene HJ. Parental awareness, habits, and social factors and their relationship to baby bottle tooth decay. Pediatr Dent 1997;19(1):22–7.[Medline]

  22. American Academy of Pediatric Dentistry. Clinical guideline on infant oral health care. Pediatr Dent 2004;26(7):67–70.[Medline]

  23. Tinanoff N, O’Sullivan DM. Early childhood caries: overview and recent findings. Pediatr Dent 1997;19(1):12–6.[Medline]





This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nishimura, M.
Right arrow Articles by Shimono, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nishimura, M.
Right arrow Articles by Shimono, T.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS