Journal of Bodywork & Movement Therapies (2014) 18, 87e91
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/jbmt
ACCURACY STUDY
Accuracy study of the main screening tools for temporomandibular disorder in children and adolescents Tatiana Oliveira de Santis, DDS a, Lara Jansiski Motta, PhD b, Daniela Aparecida Biasotto-Gonzalez, PhD a, Raquel Agnelli Mesquita-Ferrari, PhD a, Kristianne Porta Santos Fernandes, PhD a, Camila Haddad Leal de Godoy, DDS a, Thays Almeida Alfaya, DDS c, Sandra Kalil Bussadori, PhD a,* a
Rehabilitation Sciences Post Graduation Program, University of Nove de Julho, Sa˜o Paulo, SP, Brazil Pediatric Dentistry, University of Nove de Julho-Sa˜o Paulo, SP, Brazil c Dental Clinic Post Graduation Program, University Federal Fluminense, Nitero´i, RJ, Brazil b
Received 22 February 2013; received in revised form 25 May 2013; accepted 26 May 2013
KEYWORDS Temporomandibular joint disorders; Facial pain; Child
Summary The aims of the present study were to assess the degree of sensitivity and specificity of the screening questionnaire recommended by the American Academy of Orofacial Pain (AAOP) and the patient-history index proposed by Helkimo (modified by Fonseca) and correlate the findings with a clinical exam. All participants answered the questionnaires and were submitted to a clinical exam by a dentist who had undergone calibration training. Both the AAOP questionnaire and Helkimo index achieved low degrees of sensitivity for the detection of temporomandibular disorder (TMD), but exhibited a high degree of specificity. With regard to concordance, the AAOP questionnaire and Helkimo index both achieved low levels of agreement with the clinical exam. The different instruments available in the literature for the assessment of TMD and examined herein exhibit low sensitivity and high specificity when administered to children and adolescents stemming from difficulties in comprehension due to the age group studied and the language used in the self-explanatory questions. ª 2013 Elsevier Ltd. All rights reserved.
* Corresponding author. R. Vergueiro, 235/249 e Liberdade Sa ˜o Paulo, 01504-001, Brazil. Tel.: þ55 11 33859222. E-mail address: [email protected] (S.K. Bussadori). 1360-8592/$ - see front matter ª 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jbmt.2013.05.018
88
Introduction Temporomandibular disorder (TDM) is a term employed for functional alterations related to the temporomandibular joint, muscles of mastication and associated structures (Ebrahimi et al., 2011). Joint sounds, limited range of motion or deviation during the function of the mandible, pain, facial deformities and headache are among the symptoms of this disorder (Catanzariti et al., 2005; Velez et al., 2007). Approximately 34.7% of adolescents are affected by TMD (Ebrahimi et al., 2011), whereas the prevalence among children is around 23.7% (Tecco et al., 2011). Considering the multifactorial aetiology of this disorder, a number of neuromuscular, psychological and anatomic aspects should be evaluated for the establishment of the diagnosis (Okeson, 2008). Different assessment tools have been employed for the assessment of TMD, such as questionnaires (Manfredi et al., 2001), patient-history indices (Bevilaqua-Grossi et al., 2006; Fonseca et al., 1994), clinical indices and diagnostic criteria (Cavalcanti et al., 2010; de Lucena et al., 2006; Manfredini et al., 2011). Evaluation methods allow the standardized classification of the severity of the disorder and the categorisation of signs and symptoms so that the diagnosis can be properly established (Manfredini et al., 2006; Miller et al., 2000; Perillo et al., 2011). The aims of the present study were to assess the degree of sensitivity and specificity of the screening questionnaire for orofacial pain and TMD recommended by the American Academy of Orofacial Pain (AAOP) (Manfredi et al., 2001) and the patient-history index proposed by Helkimo (modified by Fonseca et al., 1994) for individuals between six and 18 years of age and correlate the findings with a specific clinical exam for the diagnosis of TMD using the Research Diagnostic Criteria for Temporomandibular Disorders (RDC/ TMD) (Manfredini et al., 2006).
Methods A cross-sectional study was carried out with individuals between six and 18 years of age enrolled at the Instituto Rogacionista in Sao Paulo, SP, Brazil. The inclusion criteria were age between six and 18 years, presence of the 2nd primary molar and 1st permanent molar in participants between six and 10 years of age and the presence of the 1st permanent molar in participants between 11 and 18 years of age. Individuals under medical, psychological or dental treatment and those with dentofacial deformities were excluded. This study was carried out in compliance with the norms that regulate studies involving human subjects contained in Resolutions n 196/96 and 251/97 of the Brazilian National Health Council and was approved under process number n 233,931/2009. The guardians of the children and adolescents signed statements of informed consent. The dentist underwent a calibration and training exercise. A researcher experienced with the administration of the questionnaires served as the gold standard in
T.O. de Santis et al. this phase. The training model proposed by the International Consortium for RDC/TMD-Based Research was used for the clinical exam, which was performed on five participants on three separate occasions to obtain a standard and avoid errors on the part of the examiner. All participants answered the AAOP questionnaire, which is composed of 10 self-explanatory questions with “yes” and “no” answers on the most frequent signs and symptoms of orofacial pain and TMD (Appendix 1), and the Helkimo patient-history index (modified by Fonseca) (Appendix 2), which is made up of 10 self-explanatory questions with “yes” and “no” answers based on different symptoms of masticatory dysfunction (subjective symptoms). The participants were then submitted a clinical exam by a single calibrated examiner (dentist) using the Research Diagnostic Criteria for Temporomandibular Disorders (RDC/ TMD). This measure is divided into two axes (I and II). Axis I is used to identify the complex interaction between physical and psychological aspects of chronic pain and allows a reliable measure of signs and symptoms of TMD through a clinical exam. Axis II determines associated psychological and psychosocial factors through the administration of a questionnaire. In the present study, the participants were only submitted to the clinical exam (Axis I), which consisted of an extra-oral and intra-oral inspection of the teeth and occlusion, palpation of the sternocleidomastoid, temporal, masseter, digastric and medial pterygoid muscles of the temporomandibular joint and an analysis of mandibular range of motion with the use of a digital calliper (Mytutoio) for the measurement of maximal mouth opening and lateral movements. The following were also analysed: joint sounds (clicking in temporomandibular joint upon opening and/or closing the mouth); dental wear (occlusal or incisal) indicative of possible parafunctional habits, such as teeth grinding; direct report regarding individual stress status; recent history of microtrauma in the orofacial region; and the investigation of frequent headache, facial pain, fatigue/ difficulty during mastication, bruxism, psychological aspects of the child, digit sucking, pacifier sucking and nail biting. The data were organized into tables and graphs and statistically treated. Frequencies and percentages were calculated for the results of the AAOP questionnaire, Helkimo Index (modified by Fonseca) and clinical exam regarding the presence or absence of TMD. Contingency tables were constructed for the determination of sensitivity, specificity, positive predictive value and negative predictive value, considering the clinical exam as the gold standard. Using these same tables, Kappa concordance coefficients were determined. All analyses were performed using the SAS program for Windows, v.9.2.
Results Among the 110 children analysed, 51 (46.3%) were female and 59 (53.6%) were male. Mean age was 8.18 years. Table 1 displays the frequency and percentage of patients with and without TMD, as determined by the AAOP
Accuracy study of TMD screening tools
89
Table 1 Frequency and percentage of patients with and without TMD as determined by the AAOP questionnaire, Helkimo index (modified by Fonseca) and the clinical exam using the Research Diagnostic Criteria (RDC).
Table 3 Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of assessment tools analysed. Variables
Sensitivity
Specificity
PPV
NPV
Variables
Frequency
Percentage
AAOP questionnaire With TMD Without TMD
48.86% 53.40%
77.27% 77.27%
28.33% 26.56%
90.00% 89.13%
60 50
54.55% 45.45%
AAOP Helkimo index (modified by Fonseca)
Helkimo index (modified by Fonseca) With TMD 64 Without TMD 46 Clinical exam (RDC) With TMD Without TMD
58.18% 41.82%
22 88
20.00% 80.00%
questionnaire, Helkimo index (modified by Fonseca) and the clinical exam using the RDC/TMD. No statistically significant differences were found between genders regarding the presence of TMD or the answers given on the questionnaires (p > 0.05). Considering the clinical exam as the gold standard, contingency tables were constructed, crossing information on the presence and absence of TMD with the AAOP questionnaire and Helkimo Index (Table 2). Table 3 displays the results obtained from the assessment tools with regard to sensitivity, specificity, positive predictive value and negative predictive value. Sensitivity is the capacity to recognize true positives (proportion of affected individuals determined by the test in relation to the total number of true positives). The AAOP and Helkimo index recognized 43/88 (48.86%) and 47/88 (53.40%) of the true positives, respectively. Specificity regards the capacity to recognize true negatives (proportion of non-affected individuals determined by the test in relation to the total number of true negatives). Both questionnaires identified 17/22 of the true negatives and therefore achieved the same degree of specificity (77.27%).
Table 2 Contingency tables crossing information on the presence and absence of TMD with the results of the AAOP questionnaire and Helkimo Index. Variables
Clinical exam With TMD
AAOP questionnaire With TMD 17 Without TMD 5 Total 22
Total Without 43 45 88
60 50 110
Helkimo index (modified by Fonseca) With TMD 17 47 Without TMD 5 41 Total 22 88
64 46 110
Sensitivity and specificity are inherent to any assessment tool. However, predictive values are calculated to determine validity in accordance with the prevalence of the health condition measured. The positive predictive value is the likelihood that a positive case identified by the measure is truly positive (number of true positives on the test over the total number of cases considered true by the gold standard exam). The AAOP and Helkimo index achieved positive predictive values of 28.33% (17/ 60) and 26.56% (17/64), respectively. The negative predictive value is the likelihood that a negative case identified by the measure is truly negative (number of true negatives on the test over the total number of cases considered negative by the gold standard exam). The AAOP and Helkimo index achieved negative predictive values of 90.0% (45/50) and 89.13% (41/46), respectively. Analysing the results, the AAOP questionnaire and Helkimo Index demonstrated low degrees of sensitivity, but high degrees of specificity regarding the detection of TMD in children and adolescents. A low positive predictive value denotes low sensitivity, whereas a better negative predictive value denotes high specificity. It can therefore be said that these tools better assess the absence of TMD when the population analysed is made up of children and adolescents. With regard to concordance, the AAOP questionnaire and Helkimo index both achieved low levels of agreement with the clinical exam, as demonstrated by the low degree of sensitivity of the instruments.
Discussion The present findings demonstrate that the assessment tools used for the diagnosis of TMD exhibit low sensitivity and high specificity when applied to children and adolescents. Difficulties in comprehension by individuals in the age group studied and the language employed on the selfexplanatory questionnaires may have been the main reasons for the results obtained in the present investigation. A number of studies have sought to assess TMD in children and adolescents by analysing characteristics common to the disorder in this age group (Casanova-Rosado et al., 2006; List et al., 2001; Nilsson, 2007; Nilsson et al., 2009). The present study analysed the correlation between clinical findings and results obtained from assessment tools.
90 The Helkimo patient-history index modified by Fonseca et al. (1994) has been employed in a number of studies for the classification of individuals based on symptoms of TMD (Bonjardim et al., 2009; de Oliveira et al., 2006; Nomura et al., 2007). The simplicity of this index favours its use in population-based epidemiological studies. However, it has not been completely validated and does not offer a diagnostic classification of TMD. Another limitation regards its scoring system, as three affirmative answers to the questions on headache, neck pain and the perception of emotional tension classify the volunteer as having mild TMD. However, these same symptoms can occur in an isolated manner without any association with this disorder. In the present study, the Helkimo patient-history index modified by Fonseca exhibited a low degree of sensitivity (53.40%), which denotes a low capacity for recognising true positives, and a high degree of specificity (77.27%), which is the power to distinguish true negatives, in comparison to the clinical exam for the detection of TMD in children and adolescents. Manfredi et al. (2001) administered the orofacial pain questionnaire recommended by the AAOP to 46 adults with TMD and found sensitivity and specificity rates of 85.37% and 80%, respectively, for patients with myogenous TMD. However, low degrees of sensitivity and specificity were found for joint disorders. The authors concluded that the questionnaire is useful for the prescreening of patients, but does not allow a definitive diagnosis. In the present study, low sensitivity (48.86%) and high specificity (77.27%) were found in the detection of TMD in children and adolescents. The disagreement between the findings in the present investigation and the study cited may stem from difficulties in comprehension on the part of young patients due to the specific language of the questions. However, analysing the type of patient each of the assessment tools identifies, the AAOP questionnaire requires pain during function for every subtype of muscle-related TMD, whereas the RDC/ TMD does not have the same requirement (Reiter et al., 2012). The RDC/TMD is one of the few measures to define operational criteria for the diagnosis of TMD. Moreover, its proven psychometric properties and accuracy make it one of the most appropriate tools for the assessment of this disorder (Manfredini et al., 2006). In the present study, the decision was made not to employ the questions on this measure due to the fact that it was specifically designed for adults and addresses subjects that children would be unable to discuss due to their young age. Thus, only Axis I was administered to the sample. However, studies employing Axis II or both axes have demonstrated the extent to which psychological factors are found in young individuals (Casanova-Rosado et al., 2006; List et al., 2001). In conclusion, the AAOP and Helkimo Index (modified by Fonseca) exhibit low sensitivity and high specificity when administered to children and adolescents, which may be attributed to difficulties in comprehension due to the age group studied and the language used in the self-explanatory questions.
T.O. de Santis et al.
Appendix 1. Screening questionnaire for TMD recommended by the American Academy of Orofacial Pain.
History
Yes No
1. Do you have difficulty, pain or both when opening your mouth, for instance, to yawn? 2. Does your jaw get “stuck”, “locked” or out of place? 3. Do you have difficulty, pain or both when chewing, speaking or using your jaws? 4. Do you hear sounds in your jaw joints? 5. Do your jaws get stiff, tight or tired regularly? 6. Do you have pain in or around your ears, temples or cheeks? a-() headache; b-() neck pain; c-() toothache 7. Do you get headaches, neck pain or toothaches often? 8. Have you suffered recent trauma to your head, neck or jaws? 9. Have you felt any recent change in your bite? 10. Have you had recent treatment for an unexplained problem in your jaw joints?
Appendix 2. Fonseca patient-history questionnaire and index.
History 1. Do you have trouble opening your mouth? 2. Do you have trouble moving your mandible from side to side? 3. Do you feel discomfort or muscle pain when chewing? 4. Do you get headaches often? 5. Do you feel neck and/or shoulder pain? 6. Do you feel pain in or around your ears? 7. Do you hear any sounds in your TMJ? 8. Do you consider your bite to be “normal”? 9. Do you only use one side of your mouth to chew? 10. Do you feel pain in your face when you wake up?
Yes
No
Sometimes
Accuracy study of TMD screening tools
References Bevilaqua-Grossi, D., Chaves, T.C., de Oliveira, A.S., MonteiroPedro, V., 2006. Anamnestic index severity and signs and symptoms of TMD. Cranio 24 (2), 112e118. Bonjardim, L.R., Lopes-Filho, R.J., Amado, G., Albuquerque Jr., R.L., Goncalves, S.R., 2009. Association between symptoms of temporomandibular disorders and gender, morphological occlusion, and psychological factors in a group of university students. Indian J. Dent Res. 20 (2), 190e194. Casanova-Rosado, J.F., Medina-Solis, C.E., Vallejos-Sanchez, A.A., Casanova-Rosado, A.J., Hernandez-Prado, B., Avila-Burgos, L., 2006. Prevalence and associated factors for temporomandibular disorders in a group of Mexican adolescents and youth adults. Clin. Oral Investig. 10 (1), 42e49. Catanzariti, J.F., Debuse, T., Duquesnoy, B., 2005. Chronic neck pain and masticatory dysfunction. Jt. Bone Spine 72 (6), 515e519. Cavalcanti, R.F., Studart, L.M., Kosminsky, M., de Goes, P.S., 2010. Validation of the multimedia version of the RDC/TMD axis II questionnaire in Portuguese. J. Appl. Oral Sci. 18 (3), 231e236. de Lucena, L.B., Kosminsky, M., da Costa, L.J., de Goes, P.S., 2006. Validation of the Portuguese version of the RDC/TMD Axis II questionnaire. Braz. Oral Res. 20 (4), 312e317. de Oliveira, A.S., Dias, E.M., Contato, R.G., Berzin, F., 2006. Prevalence study of signs and symptoms of temporomandibular disorder in Brazilian college students. Braz. Oral Res. 20 (1), 3e7. Ebrahimi, M., Dashti, H., Mehrabkhani, M., Arghavani, M., Daneshvar-Mozafari, A., 2011. Temporomandibular disorders and related factors in a group of Iranian adolescents: a crosssectional survey. J. Dent Res. Dent Clin. Dent Prospects 5 (4), 123e127. Fonseca, D.M., Bonfate, G., Valle, A.L., Freitas, S.F.T., 1994. Diagno ¸a ´stico pela anamnese da disfunc ˜o craniomandibular. Rev. Gaucha Odontol. 42 (1), 23e28. List, T., Wahlund, K., Larsson, B., 2001. Psychosocial functioning and dental factors in adolescents with temporomandibular disorders: a case-control study. J. Orofac Pain 15 (3), 218e227. Manfredi, A.P.S., Silva, A.A., Vendite, L.L., 2001. The sensibility appreciation of the questionnaire for selection of orofacial pain and temporomandibular disorders recommended by the American Academy of orofacial Pain. Rev. Bras Otorrinolaringol 67 (6), 763e768.
91 Manfredini, D., Chiappe, G., Bosco, M., 2006. Research diagnostic criteria for temporomandibular disorders (RDC/TMD) axis I diagnoses in an Italian patient population. J. Oral Rehabil. 33 (8), 551e558. Manfredini, D., Guarda-Nardini, L., Winocur, E., Piccotti, F., Ahlberg, J., Lobbezoo, F., 2011. Research diagnostic criteria for temporomandibular disorders: a systematic review of axis I epidemiologic findings. Oral Surg. Oral Med. Oral Pathol. Oral Radiol. Endod. 112 (4), 453e462. Miller, V.J., Karic, V.V., Myers, S.L., Exner, H.V., 2000. The temporomandibular opening index (TOI) in patients with closed lock and a control group with no temporomandibular disorders (TMD): an initial study. J. Oral Rehabil. 27 (9), 815e816. Nilsson, I.M., 2007. Reliability, validity, incidence and impact of temporormandibular pain disorders in adolescents. Swed Dent J. Suppl. 183, 7e86. Nilsson, I.M., Drangsholt, M., List, T., 2009. Impact of temporomandibular disorder pain in adolescents: differences by age and gender. J. Orofac Pain 23 (2), 115e122. Nomura, K., Vitti, M., Oliveira, A.S., Chaves, T.C., Semprini, S., Sie ´ssere, S., Hallak, J.E.C., Regalo, S.C.H., 2007. Use of the Fonseca’s questionnaire to assess the prevalence and severity of temporomandibular disorders in Brazilian dental undergraduates. Braz. Dent J. 18 (2), 163e167. Okeson, J.P., 2008. Mangement of Temporomandibular Disorders and Occlusion. Mosby, Philadelphia. Perillo, L., Femminella, B., Farronato, D., Baccetti, T., Contardo, L., Perinetti, G., 2011. Do malocclusion and Helkimo index >/Z 5 correlate with body posture? J. Oral Rehabil. 38 (4), 242e252. Reiter, S., Goldsmith, C., Emodi-Perlman, A., Friedman-Rubin, P., Winocur, E., 2012. Masticatory muscle disorders diagnostic criteria: the American Academy of orofacial pain versus the research diagnostic criteria/temporomandibular disorders (RDC/TMD). J. Oral Rehabil. 39 (12), 941e947. Tecco, S., Crincoli, V., Di Bisceglie, B., Saccucci, M., Macrl, M., Polimeni, A., Festa, F., 2011. Signs and symptoms of temporomandibular joint disorders in Caucasian children and adolescents. Cranio 29 (1), 71e79. Velez, A.L., Restrepo, C.C., Pelaez-Vargas, A., Gallego, G.J., Alvarez, E., Tamayo, V., Tamayo, M., 2007. Head posture and dental wear evaluation of bruxist children with primary teeth. J. Oral Rehabil. 34 (9), 663e670.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/jbmt
ACCURACY STUDY
Accuracy study of the main screening tools for temporomandibular disorder in children and adolescents Tatiana Oliveira de Santis, DDS a, Lara Jansiski Motta, PhD b, Daniela Aparecida Biasotto-Gonzalez, PhD a, Raquel Agnelli Mesquita-Ferrari, PhD a, Kristianne Porta Santos Fernandes, PhD a, Camila Haddad Leal de Godoy, DDS a, Thays Almeida Alfaya, DDS c, Sandra Kalil Bussadori, PhD a,* a
Rehabilitation Sciences Post Graduation Program, University of Nove de Julho, Sa˜o Paulo, SP, Brazil Pediatric Dentistry, University of Nove de Julho-Sa˜o Paulo, SP, Brazil c Dental Clinic Post Graduation Program, University Federal Fluminense, Nitero´i, RJ, Brazil b
Received 22 February 2013; received in revised form 25 May 2013; accepted 26 May 2013
KEYWORDS Temporomandibular joint disorders; Facial pain; Child
Summary The aims of the present study were to assess the degree of sensitivity and specificity of the screening questionnaire recommended by the American Academy of Orofacial Pain (AAOP) and the patient-history index proposed by Helkimo (modified by Fonseca) and correlate the findings with a clinical exam. All participants answered the questionnaires and were submitted to a clinical exam by a dentist who had undergone calibration training. Both the AAOP questionnaire and Helkimo index achieved low degrees of sensitivity for the detection of temporomandibular disorder (TMD), but exhibited a high degree of specificity. With regard to concordance, the AAOP questionnaire and Helkimo index both achieved low levels of agreement with the clinical exam. The different instruments available in the literature for the assessment of TMD and examined herein exhibit low sensitivity and high specificity when administered to children and adolescents stemming from difficulties in comprehension due to the age group studied and the language used in the self-explanatory questions. ª 2013 Elsevier Ltd. All rights reserved.
* Corresponding author. R. Vergueiro, 235/249 e Liberdade Sa ˜o Paulo, 01504-001, Brazil. Tel.: þ55 11 33859222. E-mail address: [email protected] (S.K. Bussadori). 1360-8592/$ - see front matter ª 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jbmt.2013.05.018
88
Introduction Temporomandibular disorder (TDM) is a term employed for functional alterations related to the temporomandibular joint, muscles of mastication and associated structures (Ebrahimi et al., 2011). Joint sounds, limited range of motion or deviation during the function of the mandible, pain, facial deformities and headache are among the symptoms of this disorder (Catanzariti et al., 2005; Velez et al., 2007). Approximately 34.7% of adolescents are affected by TMD (Ebrahimi et al., 2011), whereas the prevalence among children is around 23.7% (Tecco et al., 2011). Considering the multifactorial aetiology of this disorder, a number of neuromuscular, psychological and anatomic aspects should be evaluated for the establishment of the diagnosis (Okeson, 2008). Different assessment tools have been employed for the assessment of TMD, such as questionnaires (Manfredi et al., 2001), patient-history indices (Bevilaqua-Grossi et al., 2006; Fonseca et al., 1994), clinical indices and diagnostic criteria (Cavalcanti et al., 2010; de Lucena et al., 2006; Manfredini et al., 2011). Evaluation methods allow the standardized classification of the severity of the disorder and the categorisation of signs and symptoms so that the diagnosis can be properly established (Manfredini et al., 2006; Miller et al., 2000; Perillo et al., 2011). The aims of the present study were to assess the degree of sensitivity and specificity of the screening questionnaire for orofacial pain and TMD recommended by the American Academy of Orofacial Pain (AAOP) (Manfredi et al., 2001) and the patient-history index proposed by Helkimo (modified by Fonseca et al., 1994) for individuals between six and 18 years of age and correlate the findings with a specific clinical exam for the diagnosis of TMD using the Research Diagnostic Criteria for Temporomandibular Disorders (RDC/ TMD) (Manfredini et al., 2006).
Methods A cross-sectional study was carried out with individuals between six and 18 years of age enrolled at the Instituto Rogacionista in Sao Paulo, SP, Brazil. The inclusion criteria were age between six and 18 years, presence of the 2nd primary molar and 1st permanent molar in participants between six and 10 years of age and the presence of the 1st permanent molar in participants between 11 and 18 years of age. Individuals under medical, psychological or dental treatment and those with dentofacial deformities were excluded. This study was carried out in compliance with the norms that regulate studies involving human subjects contained in Resolutions n 196/96 and 251/97 of the Brazilian National Health Council and was approved under process number n 233,931/2009. The guardians of the children and adolescents signed statements of informed consent. The dentist underwent a calibration and training exercise. A researcher experienced with the administration of the questionnaires served as the gold standard in
T.O. de Santis et al. this phase. The training model proposed by the International Consortium for RDC/TMD-Based Research was used for the clinical exam, which was performed on five participants on three separate occasions to obtain a standard and avoid errors on the part of the examiner. All participants answered the AAOP questionnaire, which is composed of 10 self-explanatory questions with “yes” and “no” answers on the most frequent signs and symptoms of orofacial pain and TMD (Appendix 1), and the Helkimo patient-history index (modified by Fonseca) (Appendix 2), which is made up of 10 self-explanatory questions with “yes” and “no” answers based on different symptoms of masticatory dysfunction (subjective symptoms). The participants were then submitted a clinical exam by a single calibrated examiner (dentist) using the Research Diagnostic Criteria for Temporomandibular Disorders (RDC/ TMD). This measure is divided into two axes (I and II). Axis I is used to identify the complex interaction between physical and psychological aspects of chronic pain and allows a reliable measure of signs and symptoms of TMD through a clinical exam. Axis II determines associated psychological and psychosocial factors through the administration of a questionnaire. In the present study, the participants were only submitted to the clinical exam (Axis I), which consisted of an extra-oral and intra-oral inspection of the teeth and occlusion, palpation of the sternocleidomastoid, temporal, masseter, digastric and medial pterygoid muscles of the temporomandibular joint and an analysis of mandibular range of motion with the use of a digital calliper (Mytutoio) for the measurement of maximal mouth opening and lateral movements. The following were also analysed: joint sounds (clicking in temporomandibular joint upon opening and/or closing the mouth); dental wear (occlusal or incisal) indicative of possible parafunctional habits, such as teeth grinding; direct report regarding individual stress status; recent history of microtrauma in the orofacial region; and the investigation of frequent headache, facial pain, fatigue/ difficulty during mastication, bruxism, psychological aspects of the child, digit sucking, pacifier sucking and nail biting. The data were organized into tables and graphs and statistically treated. Frequencies and percentages were calculated for the results of the AAOP questionnaire, Helkimo Index (modified by Fonseca) and clinical exam regarding the presence or absence of TMD. Contingency tables were constructed for the determination of sensitivity, specificity, positive predictive value and negative predictive value, considering the clinical exam as the gold standard. Using these same tables, Kappa concordance coefficients were determined. All analyses were performed using the SAS program for Windows, v.9.2.
Results Among the 110 children analysed, 51 (46.3%) were female and 59 (53.6%) were male. Mean age was 8.18 years. Table 1 displays the frequency and percentage of patients with and without TMD, as determined by the AAOP
Accuracy study of TMD screening tools
89
Table 1 Frequency and percentage of patients with and without TMD as determined by the AAOP questionnaire, Helkimo index (modified by Fonseca) and the clinical exam using the Research Diagnostic Criteria (RDC).
Table 3 Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of assessment tools analysed. Variables
Sensitivity
Specificity
PPV
NPV
Variables
Frequency
Percentage
AAOP questionnaire With TMD Without TMD
48.86% 53.40%
77.27% 77.27%
28.33% 26.56%
90.00% 89.13%
60 50
54.55% 45.45%
AAOP Helkimo index (modified by Fonseca)
Helkimo index (modified by Fonseca) With TMD 64 Without TMD 46 Clinical exam (RDC) With TMD Without TMD
58.18% 41.82%
22 88
20.00% 80.00%
questionnaire, Helkimo index (modified by Fonseca) and the clinical exam using the RDC/TMD. No statistically significant differences were found between genders regarding the presence of TMD or the answers given on the questionnaires (p > 0.05). Considering the clinical exam as the gold standard, contingency tables were constructed, crossing information on the presence and absence of TMD with the AAOP questionnaire and Helkimo Index (Table 2). Table 3 displays the results obtained from the assessment tools with regard to sensitivity, specificity, positive predictive value and negative predictive value. Sensitivity is the capacity to recognize true positives (proportion of affected individuals determined by the test in relation to the total number of true positives). The AAOP and Helkimo index recognized 43/88 (48.86%) and 47/88 (53.40%) of the true positives, respectively. Specificity regards the capacity to recognize true negatives (proportion of non-affected individuals determined by the test in relation to the total number of true negatives). Both questionnaires identified 17/22 of the true negatives and therefore achieved the same degree of specificity (77.27%).
Table 2 Contingency tables crossing information on the presence and absence of TMD with the results of the AAOP questionnaire and Helkimo Index. Variables
Clinical exam With TMD
AAOP questionnaire With TMD 17 Without TMD 5 Total 22
Total Without 43 45 88
60 50 110
Helkimo index (modified by Fonseca) With TMD 17 47 Without TMD 5 41 Total 22 88
64 46 110
Sensitivity and specificity are inherent to any assessment tool. However, predictive values are calculated to determine validity in accordance with the prevalence of the health condition measured. The positive predictive value is the likelihood that a positive case identified by the measure is truly positive (number of true positives on the test over the total number of cases considered true by the gold standard exam). The AAOP and Helkimo index achieved positive predictive values of 28.33% (17/ 60) and 26.56% (17/64), respectively. The negative predictive value is the likelihood that a negative case identified by the measure is truly negative (number of true negatives on the test over the total number of cases considered negative by the gold standard exam). The AAOP and Helkimo index achieved negative predictive values of 90.0% (45/50) and 89.13% (41/46), respectively. Analysing the results, the AAOP questionnaire and Helkimo Index demonstrated low degrees of sensitivity, but high degrees of specificity regarding the detection of TMD in children and adolescents. A low positive predictive value denotes low sensitivity, whereas a better negative predictive value denotes high specificity. It can therefore be said that these tools better assess the absence of TMD when the population analysed is made up of children and adolescents. With regard to concordance, the AAOP questionnaire and Helkimo index both achieved low levels of agreement with the clinical exam, as demonstrated by the low degree of sensitivity of the instruments.
Discussion The present findings demonstrate that the assessment tools used for the diagnosis of TMD exhibit low sensitivity and high specificity when applied to children and adolescents. Difficulties in comprehension by individuals in the age group studied and the language employed on the selfexplanatory questionnaires may have been the main reasons for the results obtained in the present investigation. A number of studies have sought to assess TMD in children and adolescents by analysing characteristics common to the disorder in this age group (Casanova-Rosado et al., 2006; List et al., 2001; Nilsson, 2007; Nilsson et al., 2009). The present study analysed the correlation between clinical findings and results obtained from assessment tools.
90 The Helkimo patient-history index modified by Fonseca et al. (1994) has been employed in a number of studies for the classification of individuals based on symptoms of TMD (Bonjardim et al., 2009; de Oliveira et al., 2006; Nomura et al., 2007). The simplicity of this index favours its use in population-based epidemiological studies. However, it has not been completely validated and does not offer a diagnostic classification of TMD. Another limitation regards its scoring system, as three affirmative answers to the questions on headache, neck pain and the perception of emotional tension classify the volunteer as having mild TMD. However, these same symptoms can occur in an isolated manner without any association with this disorder. In the present study, the Helkimo patient-history index modified by Fonseca exhibited a low degree of sensitivity (53.40%), which denotes a low capacity for recognising true positives, and a high degree of specificity (77.27%), which is the power to distinguish true negatives, in comparison to the clinical exam for the detection of TMD in children and adolescents. Manfredi et al. (2001) administered the orofacial pain questionnaire recommended by the AAOP to 46 adults with TMD and found sensitivity and specificity rates of 85.37% and 80%, respectively, for patients with myogenous TMD. However, low degrees of sensitivity and specificity were found for joint disorders. The authors concluded that the questionnaire is useful for the prescreening of patients, but does not allow a definitive diagnosis. In the present study, low sensitivity (48.86%) and high specificity (77.27%) were found in the detection of TMD in children and adolescents. The disagreement between the findings in the present investigation and the study cited may stem from difficulties in comprehension on the part of young patients due to the specific language of the questions. However, analysing the type of patient each of the assessment tools identifies, the AAOP questionnaire requires pain during function for every subtype of muscle-related TMD, whereas the RDC/ TMD does not have the same requirement (Reiter et al., 2012). The RDC/TMD is one of the few measures to define operational criteria for the diagnosis of TMD. Moreover, its proven psychometric properties and accuracy make it one of the most appropriate tools for the assessment of this disorder (Manfredini et al., 2006). In the present study, the decision was made not to employ the questions on this measure due to the fact that it was specifically designed for adults and addresses subjects that children would be unable to discuss due to their young age. Thus, only Axis I was administered to the sample. However, studies employing Axis II or both axes have demonstrated the extent to which psychological factors are found in young individuals (Casanova-Rosado et al., 2006; List et al., 2001). In conclusion, the AAOP and Helkimo Index (modified by Fonseca) exhibit low sensitivity and high specificity when administered to children and adolescents, which may be attributed to difficulties in comprehension due to the age group studied and the language used in the self-explanatory questions.
T.O. de Santis et al.
Appendix 1. Screening questionnaire for TMD recommended by the American Academy of Orofacial Pain.
History
Yes No
1. Do you have difficulty, pain or both when opening your mouth, for instance, to yawn? 2. Does your jaw get “stuck”, “locked” or out of place? 3. Do you have difficulty, pain or both when chewing, speaking or using your jaws? 4. Do you hear sounds in your jaw joints? 5. Do your jaws get stiff, tight or tired regularly? 6. Do you have pain in or around your ears, temples or cheeks? a-() headache; b-() neck pain; c-() toothache 7. Do you get headaches, neck pain or toothaches often? 8. Have you suffered recent trauma to your head, neck or jaws? 9. Have you felt any recent change in your bite? 10. Have you had recent treatment for an unexplained problem in your jaw joints?
Appendix 2. Fonseca patient-history questionnaire and index.
History 1. Do you have trouble opening your mouth? 2. Do you have trouble moving your mandible from side to side? 3. Do you feel discomfort or muscle pain when chewing? 4. Do you get headaches often? 5. Do you feel neck and/or shoulder pain? 6. Do you feel pain in or around your ears? 7. Do you hear any sounds in your TMJ? 8. Do you consider your bite to be “normal”? 9. Do you only use one side of your mouth to chew? 10. Do you feel pain in your face when you wake up?
Yes
No
Sometimes
Accuracy study of TMD screening tools
References Bevilaqua-Grossi, D., Chaves, T.C., de Oliveira, A.S., MonteiroPedro, V., 2006. Anamnestic index severity and signs and symptoms of TMD. Cranio 24 (2), 112e118. Bonjardim, L.R., Lopes-Filho, R.J., Amado, G., Albuquerque Jr., R.L., Goncalves, S.R., 2009. Association between symptoms of temporomandibular disorders and gender, morphological occlusion, and psychological factors in a group of university students. Indian J. Dent Res. 20 (2), 190e194. Casanova-Rosado, J.F., Medina-Solis, C.E., Vallejos-Sanchez, A.A., Casanova-Rosado, A.J., Hernandez-Prado, B., Avila-Burgos, L., 2006. Prevalence and associated factors for temporomandibular disorders in a group of Mexican adolescents and youth adults. Clin. Oral Investig. 10 (1), 42e49. Catanzariti, J.F., Debuse, T., Duquesnoy, B., 2005. Chronic neck pain and masticatory dysfunction. Jt. Bone Spine 72 (6), 515e519. Cavalcanti, R.F., Studart, L.M., Kosminsky, M., de Goes, P.S., 2010. Validation of the multimedia version of the RDC/TMD axis II questionnaire in Portuguese. J. Appl. Oral Sci. 18 (3), 231e236. de Lucena, L.B., Kosminsky, M., da Costa, L.J., de Goes, P.S., 2006. Validation of the Portuguese version of the RDC/TMD Axis II questionnaire. Braz. Oral Res. 20 (4), 312e317. de Oliveira, A.S., Dias, E.M., Contato, R.G., Berzin, F., 2006. Prevalence study of signs and symptoms of temporomandibular disorder in Brazilian college students. Braz. Oral Res. 20 (1), 3e7. Ebrahimi, M., Dashti, H., Mehrabkhani, M., Arghavani, M., Daneshvar-Mozafari, A., 2011. Temporomandibular disorders and related factors in a group of Iranian adolescents: a crosssectional survey. J. Dent Res. Dent Clin. Dent Prospects 5 (4), 123e127. Fonseca, D.M., Bonfate, G., Valle, A.L., Freitas, S.F.T., 1994. Diagno ¸a ´stico pela anamnese da disfunc ˜o craniomandibular. Rev. Gaucha Odontol. 42 (1), 23e28. List, T., Wahlund, K., Larsson, B., 2001. Psychosocial functioning and dental factors in adolescents with temporomandibular disorders: a case-control study. J. Orofac Pain 15 (3), 218e227. Manfredi, A.P.S., Silva, A.A., Vendite, L.L., 2001. The sensibility appreciation of the questionnaire for selection of orofacial pain and temporomandibular disorders recommended by the American Academy of orofacial Pain. Rev. Bras Otorrinolaringol 67 (6), 763e768.
91 Manfredini, D., Chiappe, G., Bosco, M., 2006. Research diagnostic criteria for temporomandibular disorders (RDC/TMD) axis I diagnoses in an Italian patient population. J. Oral Rehabil. 33 (8), 551e558. Manfredini, D., Guarda-Nardini, L., Winocur, E., Piccotti, F., Ahlberg, J., Lobbezoo, F., 2011. Research diagnostic criteria for temporomandibular disorders: a systematic review of axis I epidemiologic findings. Oral Surg. Oral Med. Oral Pathol. Oral Radiol. Endod. 112 (4), 453e462. Miller, V.J., Karic, V.V., Myers, S.L., Exner, H.V., 2000. The temporomandibular opening index (TOI) in patients with closed lock and a control group with no temporomandibular disorders (TMD): an initial study. J. Oral Rehabil. 27 (9), 815e816. Nilsson, I.M., 2007. Reliability, validity, incidence and impact of temporormandibular pain disorders in adolescents. Swed Dent J. Suppl. 183, 7e86. Nilsson, I.M., Drangsholt, M., List, T., 2009. Impact of temporomandibular disorder pain in adolescents: differences by age and gender. J. Orofac Pain 23 (2), 115e122. Nomura, K., Vitti, M., Oliveira, A.S., Chaves, T.C., Semprini, S., Sie ´ssere, S., Hallak, J.E.C., Regalo, S.C.H., 2007. Use of the Fonseca’s questionnaire to assess the prevalence and severity of temporomandibular disorders in Brazilian dental undergraduates. Braz. Dent J. 18 (2), 163e167. Okeson, J.P., 2008. Mangement of Temporomandibular Disorders and Occlusion. Mosby, Philadelphia. Perillo, L., Femminella, B., Farronato, D., Baccetti, T., Contardo, L., Perinetti, G., 2011. Do malocclusion and Helkimo index >/Z 5 correlate with body posture? J. Oral Rehabil. 38 (4), 242e252. Reiter, S., Goldsmith, C., Emodi-Perlman, A., Friedman-Rubin, P., Winocur, E., 2012. Masticatory muscle disorders diagnostic criteria: the American Academy of orofacial pain versus the research diagnostic criteria/temporomandibular disorders (RDC/TMD). J. Oral Rehabil. 39 (12), 941e947. Tecco, S., Crincoli, V., Di Bisceglie, B., Saccucci, M., Macrl, M., Polimeni, A., Festa, F., 2011. Signs and symptoms of temporomandibular joint disorders in Caucasian children and adolescents. Cranio 29 (1), 71e79. Velez, A.L., Restrepo, C.C., Pelaez-Vargas, A., Gallego, G.J., Alvarez, E., Tamayo, V., Tamayo, M., 2007. Head posture and dental wear evaluation of bruxist children with primary teeth. J. Oral Rehabil. 34 (9), 663e670.
