ISSN 1413-3555 Rev. bras. fisioter., São Carlos, v. 11, n. 5, p. 411-17, Sept./Oct. 2007 ©Revista Brasileira de Fisioterapia ReliaBility OF phOtOgRammetRy in RelatiOn tO gOniOmetRy FOR pOStuRal lOweR limB aSSeSSment Sacco IcN, alIbert S, QueIroz bWc, PrIPaS D, KIelINg I, KImura aa, Sellmer ae, malveStIo ra & Sera mt human motion and posture Biomechanics laboratory, Department of physical therapy, Speech therapy and Occupational therapy, School of medicine, universidade de São paulo, São paulo, Sp - Brazil Correspondence to: isabel de C. n. Sacco, Departamento de Fisioterapia, Fonoaudiologia e terapia Ocupacional, Faculdade de medicina, universidade de São paulo, Rua Cipotânia, 51, Cidade universitária, Cep 05360-000, São paulo, Sp – Brasil, e-mail: icnsacco@usp.br Received: 08/03/2007 - Revised: 30/06/2007 - accepted: 21/08/2007 AbstrAct Background: postural assessment and joint range-of-motion measurements are fundamental in diagnosing, planning and following up the evolution and results from physical therapy treatment. these can be done with the aid of goniometry – the most common method in physical therapy practice – and also, through technological advances, by means of photogrammetry. Objective: to investigate the parallel reliability of computerized photogrammetry, using two software tools (Corel Draw and Sapo), in rela- tion to goniometry, in four angles of the lower limbs. method: twenty-six asymptomatic volunteers of both sexes, aged between 18 and 45 years, were studied. none of them had leg length discrepancy greater than 1 cm. the tibiotarsal angle (tt), knee flexion/extension angle (F/e), quadriceps angle (Q) and subtalar angle (S) were measured. the measurement was done first with a manual goniometer and then with digital photogrammetry by means of the Corel Draw v. 12 and Sapo v.0.63 software. Re- sults: there were no statistical differences between the three evaluation methods for the tt (p= 0.9991), S (p= 0.2159) and F/e (p= 0.4027) angles. however, for the Q angle there was a significant difference between goniometry and the software used in photogrammetry (p= 0.0067), although there was no significant difference between two software tools (p= 0.9920). this showed that the photogrammetry results were not influenced by the software used. Conclusion: in these healthy young subjects, com- puterized photogrammetry showed good parallel reliability in comparison with goniometry, for all the angles evaluated except for the Q angle. therefore, in physical therapy practice, caution is needed in using Q angle measurements coming from different postural assessment methods. Key words: goniometry; photogrammetry; posture. resumo Confiabilidade da Fotogrametria em Relação a Goniometria para Avaliação Postural de Membros Inferiores Contextualização: a avaliação postural bem como a mensuração da amplitude articular são instrumentos fundamentais para o diagnóstico, planejamento e acompanhamento da evolução e dos resultados de um tratamento fisioterapêutico. elas podem ser realizadas tanto pela goniometria - método mais utilizado na clínica fisioterapêutica – como, com o avanço tecnológico, pela fotogrametria. Objetivos: Verificar a confiabilidade paralela da fotogrametria computadorizada, utilizando dois softwares, o Corel Draw e o Sapo, em relação à goniometria para quatro ângulos nos membros inferiores. Casuística e métodos: Foram estudados 26 voluntários de ambos os sexos, assintomáticos, com idade entre 18 e 45 anos, sem anisomelia de membros inferiores maior que 1 cm. Foram mensurados os ângulos tíbio-társico (tt), de flexo/extensão do joelho (flex/ext), ângulo Q (Q) e ângulo do retropé, inicialmente, com um goniômetro manual e, posteriormente, pela fotogrametria digital por meio dos softwares Corel Draw v. 12 e Sapo v.0.63. Resultados: Os ângulos tt (p= 0,9991), do retropé (p= 0,2159) e de flexo/extensão do joelho (p= 0,4027) não foram estatisticamente diferentes entre os 3 métodos de avaliação. Já o ângulo Q foi significativamente diferente entre a goniometria e os dois softwares usados na fotogrametria (p= 0,0067), embora os valores obtidos pelos mesmos não tenham diferido entre si (p= 0,9920), demonstrando que os resultados da fotogrametria não foram influenciados pelos softwares utilizados. Conclusão: para os ângulos avaliados em sujeitos jovens assintomáticos, a fotogrametria computadorizada é confiável paralelamente à goniometria, exceto para o ângulo Q. portanto, na clínica fisioterapêutica, deve-se ter cautela no uso de medidas de ângulo Q provenientes de diferentes métodos de avaliação postural. Palavras-chave: goniometria; fotogrametria; postura. Sacco iCn, alibert S, Queiroz BwC, pripas D, Kieling i, Kimura aa, et al. Rev. bras. fisioter.412 INTRODUCTION human posture is the kinematic relationship between the positions of the body joints at a given moment. in an ideal skeletal alignment, the muscles, joints and skeletal structures are expected to be in a state of dynamical balance, generating a minimal amount of effort and overload, leading to an optimal efficiency of the locomotive system1. Both postural assessment and objective joint range-of-motion measurement are of fundamental importance to diagnosis, planning and follow-up of the evolution and results of physical therapy treatment1,2. although it is widely accepted that a balanced posture is important for good functioning of the musculoskeletal structures, postural assessment is a complex phenomenon and difficult to measure3. accord- ing to iunes et al.3, this could explain why there are so few study results that can associate postural alterations to injuries or specific musculoskeletal dysfunctions. there- fore, it is important to establish trustworthy and reliable methods aimed at quantifying variables that aid postural assessment, contributing to an evidence-based develop- ment of physical therapy. manual goniometry is widely used in clinical physical therapy to assess range-of-motion4-6. For postural assess- ment, the goniometer can also be used to measure joint angles4-7. Some of the advantages of this methodology are the low-cost of the instrument and ease of measurement, which depends almost exclusively on the assessor’s previous experience4. these advantages make manual goniometry very accessible to clinical physical therapy. as a measur- ing instrument for the upper and lower limb joints, the reliability of the universal goniometer is considered good to excellent, but for measuring the range-of-motion of the trunk, its reliability is low8. Studies have shown high reliability for measures of shoulder and knee range-of- motion, when compared to methods of visual estimation and radiography, respectively9,10, and moderate reliability for ankle dorsiflexion, when compared to the reliability of the digital inclinometer11. high correlations were also found between goniometric and radiographic measures10 and between goniometry and isokinetic dynamometry measures12 for knee range-of-motion, as well as good to excellent reproducibility for shoulder range-of-motion measurements9. with the advent of technology, digital photogram- metry is now considered an alternative to quantitative assessment of asymmetries in the postural assessment, as it can be used for linear and angular measurements3,13,14. according to the american Society for photogrammetry and Remote Sensing, photogrammetry is the art, science, and technology of obtaining reliable information on physical objects and their surroundings by recording, measuring and interpreting photographic images and radiant elec- tromagnetic energy patterns, as well as other sources3,14. photogrammetry allows the recording of subtle changes and the inter-relations between different parts of the hu- man body that are difficult to measure or record by other means13,15. the use of photogrammetry may facilitate the quanti- fication of the morphological variables related to posture, bringing more reliable data than the ones obtained through visual observation. this fact is important for the credibility of clinical physical therapy, as well as for the reliability of rehabilitation research3. Furthermore, the filing process in photogrammetry has the convenience of saving space as well as time accessing recorded files. another advantage of digital photography is the possibility of conjugation with computerized mea- suring processes, resulting in computerized photogram- metry16. therefore, computerized photogrammetry is the combination of digital photography and software such as Corel Draw17-19, that allows the measurement of horizontal and vertical distances and angles for various ends, or others specifically developed for postural assessment such as Sapo (Software for assessment of posture), a nationally-funded free software developed with scientific basis, database, and web access20, 21. the measurement of the range-of-motion of joints and other parts of the body in relation to one another must be reliable and standardized, allowing not only the comparison of phases and the assessment of treatment effectiveness, but also the publication of results for the benefit of other professionals2,16,22. in a study by iunes et al.3, computerized photogram- metry in postural assessments presented acceptable inter- and intra-assessor reliability (intraclass correlation index [iCC] between 0.71 and 0.79) for most of the assessed angular measures, being therefore recommended for asymmetry and postural deviation assessments. however the repeatability of the method is low, and follow-up of pre- and post-treat- ment results may not be sufficiently reliable3. Zonnenberg et al.23 found a high inter- and intra-assessor reliability for all the angular measures taken with photogrammetry but, as with iunes et al.3, the method’s repeatability was low. in contrast, Braun and amundson24 found adequate reli- ability and repeatability for the postural assessment of the head and shoulders by means of photogrammetry. Other studies have also shown high reliability of photometric techniques for the assessment of the shoulder and trunk range-of-motion25,26. Rothstein27 classifies the different kinds of intra-as- sessor, inter-assessor and parallel reliability. parallel reli- ability compares the values or results obtained by different instruments or tests at the same time. it is used when the aim is to obtain alternative instruments, similar to the ref- erence instrument. to analyze the parallel reliability of a new instrument, it must be compared to a reference in- strument which has been previously tested and considered 413 reliable28,29. given that manual goniometry is the most common method in physical therapy practice with good to excellent reliability2,8, it is considered the reference instrument to which new methods and instruments, such as computerized photogrammetry, may be compared. in light of that, the objective of the present study was to verify the parallel reliability of computerized pho- togrammetry in relation to goniometry for four angles of the lower limbs, by means of two software tools: Corel Draw v. 12 and Sapo v. 0.63. in this sense, we endeavored to study the features of measurements that may contribute to the development of an evidence-based physical therapy assessment process. METHODS this study had a cross-sectional observational research design and was approved by the local ethics Committee (protocol number 1237/05). twenty-six volunteers of both genders (9 men and 17 women) were studied, with a total of 52 lower limbs. the inclusion criteria were asymptom- atic individuals between 18 and 45 years of age. exclusion criteria were significant postural asymmetries, leg length discrepancy (greater than 1 cm) and episodes of pain in the lower limbs and lower back during the previous three months. the subjects were asked to sign a free and informed consent, according to Resolution 196/96 of the national health Council. all subjects completed an initial questionnaire, includ- ing personal details (name, age, phone numbers, gender, and profession), and questions regarding the abovementioned exclusion criteria. the body mass and height of the subjects were measured. the tibiotarsal angle (tt), knee flexion/extension angle (F/e), quadriceps angle (Q), and subtalar angle (S) were measured by the same assessor (table 1) using manual goniometry22 and digital photogrammetry13. For all goniometric and photogrammetric measures, the indi- vidual was in the stance position on a bench (20 cm high x 40 cm long x 40 cm wide) placed 15 cm from a wall with a posture evaluation grid. two plumb lines hanged from the roof on each side of the positioning bench past the feet of the individual. an ethyl vinyl acetate (eVa) rectangle (7cm wide x 30cm long) was placed between the feet to maintain the position inter- and intra-subjects in all measurements. to replicate the goniometry of clinical practice, mea- surements were taken without marking the anatomical points. this option may affect the study results but marking ana- tomical points for goniometric measurement would be a deviation from the normal use of the instrument. Subjects wore bathing suits during both the gonio- metric and photogrammetric assessments. goniometry was performed with the subject in the stance position. the first angle to be measured was the tt by placing the fulcrum of the goniometer on the lateral malleolus. the stationary arm was projected toward the tuberosity of the distal diaphysis of the fifth metatarsus, and the moving arm toward the head of the fibula. next, the F/e angle of the knee was measured with the fulcrum on the head of the fibula, the stationary arm over the lateral surface of the thigh, projected toward the greater trochanter of the femur, and the moving arm over the fibula projected to- ward the lateral malleolus of the ankle. Both angles were measured first on the right lower limb and then on the left. after these procedures, the Q angle was measured by placing the fulcrum of the goniometer over the center of the patella, the stationary arm along the femur, projected toward the anterior superior iliac spine (aSiS), and the moving arm over the tibial tuberosity. Finally, the S angle was measured with the fulcrum of the goniometer over the midpoint between both malleoli, the stationary arm over the lower third of the tibia, and the moving arm aligned with the calcaneus. after goniometry, subjects were photographed in a private, well-lit, heated room, with a non-reflexive back- ground. anatomical points were located and marked bi- laterally with red self-adhesive tags (0.9cm in diameter) for subsequent angle calculation using the software. they were: center of the patella, tibial tuberosity and anterior superior iliac spine (aSiS) (anterior frontal plane); midpoint of the lower third of the leg, midpoint of the calcaneus, Table 1. Description of the angles measured in goniometry and photogrammetry. Angle Fulcrum Stationary arm Moving arm tibiotarsal angle (tt) lateral malleolus toward the fifth metatarsal head toward the head of fibula Knee extension/ Flexion angle (F/e) head of fibula lateral thigh surface toward greater trochanter of femur along the fíbula toward lateral malleolus Q angle * Center of the patella along the femur toward the aSiS tibial tuberosity Subtalar angle (S) midpoint between the two malleoli along the inferior third of the tibia aligned with the calcaneus * the value considered was 180º subtracted from the angle measured by the goniometer. Reliability: photogrammetry and goniometry Rev. bras. fisioter. Sacco iCn, alibert S, Queiroz BwC, pripas D, Kieling i, Kimura aa, et al. Rev. bras. fisioter.414 midpoint between the malleoli (posterior frontal plane); tuberosity of the distal diaphysis of the fifth metatarsus, lateral malleolus, head of the fibula and greater trochanter of the femur (sagittal plane). the subjects were photographed in the right and left sagittal, and anterior and posterior frontal planes with a 2 megapixel (1600 x 1200 pixel) digital camera parallel to the ground, at a distance of 3m from the bench and mounted on a level tripod 70 cm high (at knee height) (Figure 1). in the sagittal plane, the subject had the elbows flexed at 90º and, in the frontal plane, the arms were to the side of the body. the photogrammetric calculation of the angles in ques- tion was performed by means of two software programs: Corel Draw v. 12 and Sapo v. 0.63. the marked anatomical points were used to measure the angles with the axes and vortices described in the goniometry (table 1), except for the Q angle, measured directly with the extension of the line between the center of the patella and the tibial tuberosity, without the need to subtract it from 180º (Figure 2). after collection, organization and verification of the data normality by means of the Shapiro-wilks adherence test, we compared the calculated variables between the methods by using the anOVa test for repeated measures (α= 0.06) and Scheffé’s post hoc test. pearson’s correlation was also applied to the methods in order to verify the strength of the relationship between them. the pearson correlation was considered significant when the p value was less than 0.05. the r values lower than 0.40 were considered low correlation; between 0.41 and 0.59, moderate; between 0.60 and 0.79, good, and above 0.80 high. the parallel reliability tends to be lower than the intra-assessor reliability and the instrument reliability as it involves measures from different devices, and usually with different scales28. in the present study, for example, the measuring scale of the goniometer and of Corel Draw was numeric, while Sapo’s was decimal. resuLts the 26 subjects (9 males and 17 females) presented mean age of 21.7 ± 4.9 years, mean body mass of 62.7 ± 13.8 kg and mean height of 168.1 ± 11.8 cm. the tt angle (p= 0.9991), S angle (p= 0.2159) and F/e angle (p= 0.4027) were not statistically different among the three assessment methods (table 2). the Q angle was significantly different among goniometry and the two software programs used in photogrammetry (p= 0.0067) (table 2), although the values obtained by Corel Draw and Sapo did not differ from one another (p= 0.9920), showing that the software used for all the assessed angles did not influence the photogram- metry results. the tt angle presented significant correlation (p< 0.05) in all methods, with the correlation between goniometry and software near 50%, and of 85% between both photogram- metry software programs. the F/e angle presented significant correlation between goniometry and photogrammetry with Corel Draw, however a low, non-significant correlation was found between Sapo and goniometry, and between Sapo and Corel Draw. the Q angle presented good and significant correlation between Corel and goniometry, but not between goniometry and Sapo, and a high correlation between both programs. the S angle presented low, non-significant cor- relation between goniometry and the software programs. in contrast, both photogrammetry methods showed a high and significant correlation for this angle (table 3). Figure 1. Digital photo standardization (description of camera resolution, distance between camera and subject, tripod height and dimensions of the wooden stool and the rectangular eVa device placed between the feet to standardize their position). Figure 2. illustration of the angle measurements using Corel Draw v. 12. Tibiotarsal Angle Knee Flexion/extension Q Angle Subtalar Angle 415 DISCUSSION Based on these results, it was clear that goniometry and computerized photogrammetry with both software programs (Corel Draw and Sapo) were very similar for the tt, S and F/e angles. the exception was the Q angle, which was dif- ferent for goniometry and photogrammetry, but similar for both programs used in the photogrammetric calculations. although the Q angle is widely used in clinical physi- cal therapy, few studies have performed reliability tests of this measure30. it is believed that the unsatisfactory results found in the present study are due to the fact that, for this angle, anatomical reference points are distant from each other and their muscle mass and arrangement is such that hinders the positioning of the goniometer arms. Furthermore, the Q angle measure involves postures of more than one joint complex, including the pelvis, hip, femoral-patellar, and femoral-tibial complexes, adding up to almost a dozen of degrees of freedom. therefore, postural alterations in each one of the degrees of freedom of these three joint complexes (pelvis, hip and knee) may alter the measure of the Q angle both in goniometry and photogrammetry. this finding is in accordance with the studies that found a low intra- and inter-assessor reliability (iCC between 0.14 and 0.37) for the clinical measure of the Q angle, as well as a low correlation between the clinical and the radiographic measure of this angle30. in contrast, a study found iCC results over 0.80 for the intra-assessor reliability and above 0.60 for the inter-assessor reliability for the Q angle measure31. it is also argued that the lateralization of the patella can alter the Q angle measures, leading to lower values for this angle. in this case, it was proposed that the medial-lateral orientation of the patella be measured to improve the reli- ability and clinical applicability of this measure32. as for the other angles, the anatomical points are either close to the goniometer arms or on a plane on the human body, so that the arms do not need to contour anatomical irregularities. thus, for the F/e angle, for example, the side of the thigh served as reference for the positioning of the goniometer arm, aligning it with the greater trochanter of the femur. these points can be positioned directly under a goniometer arm. with regard to the pearson correlations, both software programs used in photogrammetry had a high and significant correlation. this suggests that, proportionally, the measures vary in a similar fashion, are related to one another and are reliable in parallel terms. no studies were found in the literature comparing these two software programs in postural analysis for the assessed angles. Between goniometry and Corel Draw photogrammetry, moderate and good correlations were found, with exception of the S angle, which was low. Table 2. means, standard deviations of the angular variables measured with the three assessment methods, and p values. Goniometry Corel Draw SAPo p tibiotarsal angle (degrees) 112.3 ± 4.0 112.4 ± 3.6 112.4 ± 3.4 0.9991 Subtalar angle (degrees) 7.1 ± 3.7 8.1 ± 4.5 8.1 ± 4.4 0.2159 Q angle (degrees) 15.0 ± 5.6 13.1 ± 7.8 13.1 ± 7.8 0.0068 * Knee Flexion/extension angle (degrees) 184.0 ± 4.7 181.7 ± 4.1 181.6 ± 4.3 0.4027 * represents significant difference, p< 0.05. Table 3. pearson’s Correlations among the three assessment methods, r and p values. Variables Comparisons 2 x 2 r p Correlation tibiotarsal angle gonio/Corel 0.41 0.003 moderate gonio/Sapo 0.47 0.001 moderate Corel/Sapo 0.85 0.000 high Knee Flexion/extension angle gonio/Corel 0.48 0.000 moderate gonio/Sapo 0.06 0.6721 low Corel/Sapo 0.04 0.7791 low Q angle gonio/Corel 0.65 0.0000 good gonio/Sapo 0.06 0.6721 low Corel/Sapo 0.97 0.0000 high Subtalar angle gonio/Corel -0.11 0.4682 low gonio/Sapo -0.09 0.5232 low Corel/Sapo 0.83 0.000 high Reliability: photogrammetry and goniometry Rev. bras. fisioter. Sacco iCn, alibert S, Queiroz BwC, pripas D, Kieling i, Kimura aa, et al. Rev. bras. fisioter.416 Between goniometry and the Sapo photogrammetry, a low and non-significant correlation was verified. as previ- ously described, the goniometry and the Corel Draw program scales are numeric, whereas the Sapo scale is decimal, yielding different results that may be expressed by these low correlations. CONCLUSION the study showed that, for the angles assessed in young asymptomatic individuals, computerized photogrammetry is reliable in parallel terms to goniometry, except for the Q angle. in addition to that, the measurements taken with photogrammetry, regardless of the software used, were similar and did not interfere in the assessments. therefore, in clinical physical therapy, caution should be used with Q angle measurements derived from different methods of postural assessment. REFERENCES 1. Kendall Fp, mcCreary eK, provance pg. músculos – provas e funções. 4ª ed. São paulo: editora manole; 1995. 2. harrelson gl, Swann e. medidas em reabilitação. in: andrews JR, harrelson gl, wilk Ke. Reabilitação física do atleta. 3ª ed. Rio de Janeiro: elsevier; 2005. p. 105-34. 3. iunes Dh, Castro Fa, Salgado hS, moura iC, Oliveira aS, Bevilaqua-grossi D. Confiabilidade intra e interexaminadores e repetibilidade da avaliação postural pela fotogrametria. Rev Bras Fisioter. 2005;9(3):327-34. 4. Venturni C, andré a, aguilar Bp, giacomelli B. Confiabilidade de dois métodos de avaliação da amplitude de movimento ativa de dorsiflexão do tornozelo em indivíduos saudáveis. acta Fisiatr. 2006;13(1):41-5. 5. Brosseau l, tousignant m, Budd J, Chartier n, Duciaume l, plamondon S, et al. intratester and intertester reliability and criterion validity of the parallelogram and universal goniometers for active knee flexion in healthy subjects. physiother Res int. 1997;2(3):150-66. 6. Sabari JS, maltzev i, lubarsky D, liszkay e, homel R. gonio- metric assessment of shoulder range of motion: comparison testing in supine and sitting positions. arch phys med Rehabil. 1998;79:647-51. 7. tomsich Da, nitz aJ, threlkeld aJ, Shapiro R. patellofemoral alignment: reliability. J Orthop phys ther. 1996;23(3):200-8. 8. amado-João Sm. avaliação articular. in: amado-João Sm. métodos de avaliação clínica e funcional em fisioterapia. Rio de Janeiro: guanabara Koogan; 2006. p. 39-50. 9. andrade Ja, leite Vm, teixeira-Salmela lF, araújo pmp, Juliano y. estudo comparativo entre os métodos de estima- tiva visual e goniometria para avaliação das amplitudes de movimento da articulação do ombro. acta Fisiatr. 2003;10(1): 12-6. 10. gogia pp, Braatz Jh, Rose SJ, norton BJ. Reliability and validi-ty of goniometric measurements at the knee. phys ther. 1987; 67:192-5. 11. Venturni C, andrade a, aguilar Bp, giacomelli B. Confiabili- dade de dois métodos de avaliação da amplitude de movimento ativa de dorsiflexão do tornozelo em indivíduos saudáveis. acta Fisiatr. 2006;13(1):41-5. 12. Batista lh, Camargo pR, aiello gV, Oishi J, Salvini tF. aval- iação da amplitude articular do joelho: correlação entre as me- didas realizadas com o goniômetro universal e no dinamômetro isocinético. Rev Bras Fisioter. 2006;10(2):193-8. 13. watson awS. procedure for the production of high quality photographs suitable for the recording and evaluation of pos- ture. Rev Fisioter univ São paulo. 1998;5(1):20-6. 14. aSpRS – american Society for photogrammetry and Remote Sensing. what is aSpRS – definition [homepage na internet]. Bethesda: american Society for photogrammetry and Remote Sensing; 2000 [atualizada em 2006 nov 16; acesso em 2006 Out 24]. Disponível em: http://www.asprs.org/society/about. html. 15. Cowan Dn, Jones Bh, Frykman pn, polly Jr. Dw, harman ea, Rosenstein Rm, et al. lower limb morphology and risk of overuse injury among male infantry trainees. med Sci Sports exerc. 1996;28(8):945-52. 16. watson aw, mac Donncha C. a reliable technique for the as- sessment of posture: assessment criteria for aspects of posture. J Sports med phys Fitness. 2000;40:260-70. 17. mattos F, Rodrigues al. Corel Draw 11. Rio de Janeiro: Bras- port; 2003. 18. Sacco iCn, morioka eh, gomes aa, Sartor CD, noguera gC, Onodera an, et al. implicações da antropometria para posturas sentadas em automóvel – estudos de caso. Rev Fisioter univ São paulo. 2003;10(1):34-42. 19. Sacco iCn, andrade mS, Souza pS, nisiyama m, Cantuária al, maeda Fyi, et al. método pilates em revista: aspectos biomecânicos de movimentos específicos para reestruturação postural – estudos de caso. Rev Bras Ciên e mov. 2005;13(4): 65-78. 20. portal do projeto Software para avaliação postural [homepage na internet]. São paulo: incubadora Virtual Fapesp; 2004 [atu- alizada em 2007 Jan 06; acesso em 2006 Out 24]. Disponível em: http://sapo.incubadora.fapesp.br/portal. 21. Ferreira eag. postura e controle postural: desenvolvimento e aplicação de método quantitativo de avaliação postural [tese]. São paulo: Faculdade de medicina da universidade de São paulo; 2005. 22. marques ap. manual de goniometria. 2ª ed. São paulo: manole; 2003. 23. Zonnenberg aJJ, maanen V, elvers Jwh, Oostendorp RaB. intra/interrater reliability of measurements on body posture photographs. J Cranomand pract. 1996;14(4):326-31. 24. Braun Bl, amundson lR. Quantitative assessment of head and shoulder posture. arch med phys Rehabil. 1989;70(4): 322-9. 25. Sato tO, Vieira eR, gil Coury h. análise da confiabilidade de técnicas fotogramétricas para medir a flexão anterior do tronco. Rev Bras Fisioter. 2003;7(1):53-99. 26. hayes K, walton JR, Szomor Zl, murrel gaC. Reliability of five methods for assessing shoulder range of motion. aust J physiother. 2001;47:289-94. 417 27. Rothestein Jm. measurement and clinical practice: theory and application. in: Rothstein Jm. measurement in physical therapy. new york: Churchill livingstone; 1985. 28. gadotti iC, Vieira eR, magee DJ. importance and clarification of measurement properties in rehabilitation. Rev Bras Fisioter. 2006;10(2):137-46. 29. Rodrigues Fl, Vieira eR, Benze Bg, Coury hJCg. Compa- ração entre o duplo flexímetro e o eletrogoniômetro durante o movimento de flexão anterior da coluna lombar. Rev Bras Fisioter. 2003;7(3):269-74. 30. greene CC, edwards tB, wade mR, Carson ew. Reliabil- ity of the quadriceps angle measurement. am J Knee Surg. 2001;14(2):97-103. 31. Caylor D, Fites R, worrel tw. the relationship between quad- riceps angle and anterior knee pain syndrome. J Orthop Sports phys ther. 1993;17(1):11-6. 32. herrington l, nester C. Q angle undervaluated? the relation- ship between Q angle and medio lateral position of the patella. Clin Biomech. 2004;19(10):1070-3. Reliability: photogrammetry and goniometry Rev. bras. fisioter.