FUNDAMENTO: A doença cardiovascular é a principal causa de morbi-mortalidade nos diabéticos. A isquemia do miocárdio é freqüentemente assintomática levando ao diagnóstico tardio e pior prognóstico. Sabe-se que a mulher diabética tem risco de morte cardiovascular maior em relação ao sexo masculino. OBJETIVO: Avaliar a prevalência de doença arterial coronariana (DAC) em diabéticas assintomáticas. Comparar os resultados do teste ergométrico (TE), do teste cardiopulmonar (TCP) e da cintilografia do miocárdio sob estímulo farmacológico com dipiridamol (CM) com os achados da cinecoronariografia (CINE) verificando o método de maior acurácia na identificação de DAC significativa. MÉTODOS: Foram avaliadas 104 diabéticas que realizaram TE, TCP e CM no período de dois meses da CINE. As cintilografias com MIBI-99mTc foram realizadas pela técnica de gated-SPECT. A análise estatística foi realizada pelos testes x² de Pearson e t de Student, sendo realizada, ainda, análise de regressão logística. RESULTADOS: A prevalência de DAC no grupo estudado foi de 32,7%. No TE, o teste eficaz (p=0,045), a incompetência cronotrópica (p=0,031) e o tempo de esforço realizado (p=0,022) apresentaram associação significativa com DAC. No TCP, o VO2pico e a FC atingida apresentaram associação com DAC (p=0,004 e p=0,025). A maioria das variáveis da CM mostrou importante associação com DAC (todas com p=0,001). CONCLUSÃO: Os resultados obtidos sugerem elevada prevalência de DAC em pacientes diabéticas assintomáticas, devendo ser essa uma população investigada do ponto de vista cardiovascular. Dos métodos diagnósticos não-invasivos que foram empregados, o que mostrou ter maior poder de discriminação em relação às portadoras de DAC foi a CM com dipiridamol.
ARTIGO ORIGINAL
Instituto Dante Pazzanese de Cardiologia - Universidade Federal de São Paulo, São Paulo, SP - Brasil
A prevalência de diabetes (DM) vem aumentando rápida e continuamente nas últimas décadas, tomando proporções epidêmicas1. Paradoxalmente à redução da mortalidade por doença arterial coronariana (DAC), os diabéticos vêm apresentando significativa elevação da mortalidade1. A doença cardiovascular é a principal causa de morbidade e mortalidade, acometendo cerca de metade dos diabéticos1. Nesse grupo de pacientes a isquemia do miocárdio é, freqüentemente, silenciosa, levando a diagnóstico tardio e piorando o prognóstico1.
Desde a década de 1990, estudos2-4 vêm demonstrando diferença na investigação de DAC, de acordo com o sexo do paciente, tentando atribuí-la à menor incidência e ao fato de os sintomas serem atípicos no sexo feminino. Porém, é conhecido que a mulher diabética tem risco de morte cardiovascular 7,5 vezes maior em relação à não-diabética3. Como a presença de eventos cardiovasculares (EC) pode ser modificada pela triagem apropriada, pela intervenção e pela terapêutica adequada, o diagnóstico precoce pode ser de grande benefício.
O presente estudo tem como objetivos avaliar a prevalência de DAC em mulheres diabéticas tipo 2, sem sintomas cardiovasculares, e ainda identificar qual o método diagnóstico de maior eficácia na identificação de DAC significativa no grupo já descrito, comparando o teste ergométrico (TE), o teste cardiopulmonar (TCP) e a cintilografia do miocárdio (CM) sob estímulo farmacológico com dipiridamol (dipi) com os achados da cinecoronariografia (CINE).
Inicialmente, foram incluídas 120 diabéticas tipo 2, com idade entre 40 e 80 anos, e tempo mínimo de DM de cinco anos. Excluíram-se portadoras de limitação funcional ou sintomas sugestivos de isquemia. As pacientes (P) foram convocadas (após aprovação do estudo pelo comitê de ética e pesquisa da instituição) por contato telefônico ou carta, sendo incluídas no estudo após entrevista inicial, avaliação clínica e assinatura do termo de consentimento. Após a entrevista foram excluídas 16 P (por sintomas sugestivos de isquemia ou por dificuldade de retorno para a realização dos exames). Permaneceram no estudo 104 P, sendo todas submetidas a dosagem de hemoglobina glicada (Hb1Ac), TE, TCP, CM e CINE, com intervalo máximo de dois meses e sem procedimentos de revascularização entre eles. Foram suspensos, conforme orientação prévia, os medicamentos que pudessem interferir na análise dos testes diagnósticos. Os exames foram avaliados por dois especialistas em cada área, sem nenhuma informação sobre o resultado dos demais exames.
O TE foi interpretado segundo a presença de alterações eletrocardiográficas e clínicas como sugestivo de isquemia na presença de infradesnivelamento do segmento ST (Infra ST)
Cintilografia de perfusão do miocárdio sob estímulo farmacológico com dipiridamol
A CM com dipi e em condições basais foi realizada com MIBI-99mTc segundo o protocolo padrão de dois dias. As imagens foram adquiridas pela técnica de
As imagens foram processadas pelo
A análise da função sistólica do ventrículo esquerdo (VE) foi determinada pelas informações do
Após análise estatística, observou-se que a prevalência de DAC significativa em diabéticas sem sintomas cardiovasculares foi igual a 32,7%. Das 104 P estudadas, 34 apresentaram lesão
A elevada prevalência de DAC (32,7%) encontrada no grupo estudado pode ser justificada pela característica da Instituição. Por ser um dos centros de referência na área de cardiologia do Estado de São Paulo, recebe grande número de portadores de múltiplos fatores de risco para DAC, tornando a população de alta probabilidade pré-teste (
Segundo o consenso1, publicado em 1998 pela Sociedade Americana de DM, a prevalência estimada de DAC para os portadores de DM é de 55%.
Na literatura não há trabalhos demonstrando a prevalência de DAC especificamente em mulheres portadoras de DM do tipo 2 sem sintomas cardiovasculares.
Nenhuma característica clínico-epidemiológica mostrou associação significativa com DAC no grupo estudado, pela análise univariada. Pela análise de regressão logística, entretanto, valores anormais de Hb1Ac e tabagismo foram fatores discriminantes de portadoras de DAC nessa população (
O primeiro desafio no diagnóstico de DAC nas diabéticas é determinar qual o melhor método para uma triagem inicial.
O ECG clássico em repouso adiciona pouca informação na identificação dos portadores de DAC, já que apenas exclui, ou sugere, a presença de EC agudo ou infarto do miocárdio (IM) prévio13. Com freqüência, o ECG basal em mulheres apresenta alterações da repolarização ventricular (ADRV), tornando o seu valor no diagnóstico de DAC de pouca relevância13. Sabe-se que, na fase pré-climatério, a especificidade do ECG é ainda menor, por serem mais freqüentes as ADRV relacionadas ao estrógeno, que é considerado responsável por vasoconstrição da rede coronariana resultando nas alterações mencionadas13. As ADRV são restritas ao segmento ST e à onda T, assemelhando-se às observadas durante a vigência de digitálicos que têm molécula estruturalmente semelhante ao estrógeno13. Das 104 P avaliadas, 39 (37,5%) apresentavam ADRV no ECG basal.
O TE é amplamente utilizado como método de escolha na investigação inicial de DAC. Segundo a literatura, apresenta valores de sensibilidade (S) e de especificidade (E) menores no sexo feminino em relação ao masculino, em torno de 61% e 69%, respectivamente5,8. No presente estudo, como citado anteriormente, encontramos valores inferiores.
As diabéticas, com freqüência, são ainda portadoras de insuficiência vascular periférica1. No grupo estudado, 43 P (42%) apresentaram TE ineficaz, o que talvez seja um dos motivos da baixa S encontrada. Outro inconveniente encontrado é o grande número de testes considerados "falsamente" sugestivos de isquemia pelas ADRV. Henzlova e cols.13 compararam três grupos de mulheres que realizaram a CM após esforço. Apesar das CM serem normais, aquelas que faziam reposição hormonal apresentaram 39% de TE compatíveis com isquemia, comparativamente a 22% do grupo na fase pré-menopausa e 17% do grupo pós-menopausa, ambos sem uso de estrógeno. Das 39 P (37,5%) que apresentavam ECG basal com ADRV, no grupo estudado, 10 (33,3%) apresentaram TE sugestivo de isquemia na ausência de DAC significativa, o que, provavelmente, tornou a E baixa.
O tempo de esforço realizado (Dt) pode refletir, indiretamente, a capacidade funcional das pacientes5,8. No grupo estudado, apesar da limitação de membros inferiores e do baixo Dt, observou-se associação significativa entre o "t e a presença de DAC (p=0,022). A FC atingida também mostrou associação com DAC. No grupo de P com DAC a FC foi, em média, igual a 132,45 min, e nas P sem DAC, igual a 142,07 min, com p= 0,027 (
Sabe-se que a FC aumenta linearmente durante o exercício dinâmico de acordo com o incremento da carga de trabalho e do consumo de O2. No sexo feminino observa-se, em geral, maior elevação da FC em relação ao masculino5 e sabe-se que seu aumento inadequado pode ser um sinal importante de DAC. Azarbal e cols.6 verificaram que a IC é um importante preditor de mortalidade e que deveria ser incorporada aos algoritmos de estratificação de risco pelo TE. Entretanto, os autores descreveram que, no seu estudo, a CM foi melhor preditor de mortalidade e de EC em relação à presença de IC. No presente estudo, observou-se associação significativa entre IC e DAC (p= 0,031). Das 34 P com DAC importante, 26 (73,5%) apresentaram critérios de IC.
Como dito anteriormente, o TCP pode fornecer informações valiosas a respeito da capacidade funcional. Nesse estudo, tentou-se avaliar o valor do TCP no diagnóstico de DAC em diabéticas assintomáticas. Em razão do menor custo em relação à CM e pela possibilidade da associação das variáveis do ECG e clínicas obtidas ao esforço com os parâmetros funcionais, talvez o TCP fosse um método diagnóstico não-invasivo de custo-efetividade superior ao TE e à CM com dipi.
São poucos os estudos publicados verificando a utilidade do TCP para a investigação de DAC14,15. Klainman e cols.14 demonstraram que as variáveis ventilatórias estão reduzidas no esforço quando se comparam P hígidas com portadoras de DAC. No presente estudo, observou-se que o VO2pico e a FC atingida ao esforço mostraram associação com a presença de DAC (p=0,004 e p=0,025). O valor de S encontrados para o TCP no diagnóstico de DAC também foi baixo, igual a 44,1%. Já o valor de E foi superior ao do TE (72,8%). Das 34 pacientes com DAC na CINE, apenas 15 apresentavam TCP sugestivo de isquemia. Entretanto, das 70 pacientes sem DAC significativa, 51 apresentaram o TCP não sugestivo de isquemia ao esforço.
A identificação da gravidade e da extensão dos defeitos de perfusão permite estabelecer o grau de repercussão funcional de uma obstrução coronária. Como nem sempre uma lesão anatômica da artéria coronária produz alterações isquêmicas, há situações em que uma investigação conjunta anatômica e funcional seria a conduta mais adequada.
Além da importância na estratificação de risco e no manuseio clínico, a CM é considerada importante método não-invasivo que adiciona acurácia diagnóstica ao TE para ambos os sexos16,17. Sua elevada E é particularmente útil para identificar mulheres com TE considerado "falsamente positivo"8. Os valores médios de S e E da CM com tálio-201 pela técnica plana são de 84% e 87%, respectivamente, e pela técnica de SPECT a média de S eleva-se para 88%, sem diminuir a E (87%). Com os avanços da metodologia, os valores de S e E mostraram incremento. Smanio e cols. encontraram pela técnica de
Até o presente momento, encontra-se na literatura pequeno número de estudos sobre a CM em mulheres. Taillefer e cols.19 compararam tálio-201 e MIBI-99mTc em 115 mulheres, observando ausência de diferença significativa na S entre os dois fármacos19. Porém, houve grande incremento da E com a técnica de
Estudos anteriores20,21 verificaram valor comparável para os estímulos com dipi e adenosina, em relação às imagens após o exercício. Verani e cols.20 comprovou a superioridade da CM com dipi em relação ao TE submáximo no diagnóstico de DAC. Em razão da baixa capacidade funcional verificada no grupo estudado (43% de TE ineficazes), fez-se importante investigação por método que não exigisse esforço intenso e com alto poder diagnóstico, como CM com dipi.
Poucos estudos20,25 verificam o valor da CM em diabéticos assintomáticos. Para tal finalidade, o estudo multicêntrico DIAD (
Nesse estudo, os valores de S e E da CM no diagnóstico de DAC foram elevados (94% e 97%), sendo semelhantes aos descritos em literatura16. A maioria das variáveis da CM apresentou associação com DAC (
As informações funcionais também trazem importantes contribuições no diagnóstico e no prognóstico de DAC16,28. Nesse estudo verificamos que os valores de FEVE basal e após dipi mostraram associação com DAC (p = 0,001; para ambas). Porém, a queda da FEVE após dipi em relação à basal não mostrou associação significativa com DAC (p = 0,651).
Estudo de Yamagishi e cols.28 demonstraram que a queda dos valores de FEVE após estresse promove incremento na S para detecção de doença multiarterial. Pode-se tentar explicar essa discordância da literatura pelas alterações de microcirculação comum em diabéticos e que podem causar isquemia subendocárdica e queda da FEVE durante o estímulo farmacológico na ausência de lesão obstrutiva anatomicamente grave.
As alterações de espessamento sistólico e da contratilidade também mostraram importante associação com DAC (p = 0,001). Verificou-se, ainda, que as presenças de dilatação de VE e de captação pulmonar do MIBI na fase de estresse também apresentaram associação com DAC (p = 0,017 e p=0,003). Nos portadores de disfunção de VE em repouso ou induzida pelo estresse, pode ocorrer extravasamento do radiofármaco para o espaço intersticial pulmonar, em razão do aumento da pressão diastólica final12. Adicionalmente, com o tempo circulatório prolongado, ocorre aumento da atividade radioativa nos campos pulmonares. Embora esse fenômeno seja mais freqüente na CM com tálio-201, o mesmo pode ser observado com Mibi-99mTc.
Outra variável analisada foi dilatação transitória ventricular esquerda (TID) após a infusão de dipiridamol. A presença de TID após a fase de estresse, em geral, está relacionada à falência de VE, guardando forte associação à DAC com disfunção isquêmica induzida pelo estresse, elemento de valor preditivo inquestionável na determinação de sobrevida11. Hansen e cols.12 verificaram em 1.129 P importante associação entre a presença de TID após a infusão de dipi e DAC12, atribuindo essa dilatação da cavidade ventricular a uma hipoperfusão subendocárdica difusa. No grupo estudado observou-se associação significativa entre a presença de TID e DAC (p=0,001).
Analisando-se variáveis clínicas e eletrocardiográficas durante a infusão de dipi, observou-se que DP e alterações do ECG não mostraram associação significativa com DAC (p=0,166 e 0,053, respectivamente). O infradesnivelamento de ST durante a infusão de dipi pode ser secundário ao fenômeno de "roubo de fluxo"20. Porém, nesse grupo não se observou também associação entre infradesnivelamento de ST e CC importante na CINE (p=0,467).
O método mostrou-se seguro e com paraefeitos apresentados semelhantes aos descritos em literatura21.
Como limitação do estudo, consideramos a análise qualitativa das imagens da perfusão. A interpretação pela análise quantitativa por escores descritos na literatura não é usualmente empregada em nosso serviço, bem como na maioria dos serviços de medicina nuclear clínica. Portanto, não faria sentido realizarmos a investigação de DAC por meio de metodologia que não seja empregada na prática clínica diária.
Finalmente, entendemos que esse estudo é de grande valor no manuseio clínico das diabéticas. Sabendo-se que o risco cardiovascular é alto nos portadores de DM e que mulheres têm manifestações clínicas de difícil caracterização, a avaliação cardiovascular preventiva faz-se justificada no grupo estudado. A presença de ADRV no ECG basal torna as provas de esforço de menor acurácia na identificação de DAC. Aliada à baixa capacidade funcional freqüente nas portadoras de DM, o valor dos métodos que dependem da realização de exercício vigoroso são limitados. Entretanto, não é custo-efetivo realizar investigação anatômica em todas as diabéticas.
Por conseqüência do exposto, infere-se que a realização de técnica segura e de elevado poder diagnóstico e prognóstico, tal como a CM associada ao estímulo farmacológico com dipiridamol, pode ser o método de escolha na avaliação inicial das diabéticas mesmo na ausência de sintomas cardiovasculares.
Como conclusão primária, verificou-se, por meio dos resultados obtidos, que a prevalência de DAC em diabéticas sem sintomas cardiovasculares é elevada (32,7%), devendo essa população ser investigada do ponto de vista cardiovascular. Concluiu-se, ainda, que o método diagnóstico não-invasivo que mostrou maior poder de discriminar as portadoras de DAC foi a CM sob estímulo farmacológico com dipi, devendo ser esse o método de escolha para esse grupo de pacientes.
Declaro não haver conflito de interesses pertinentes.
O presente estudo não teve fontes de financiamento externas.
Este artigo é parte de tese de doutorado de Paola Emanuela Poggio Smanio pela Universidade Federal de São Paulo-UNIFESP
1. Consensus development conference on the diagnosis of coronary heart disease in people with diabetes: 10-11 February 1998, Miami, Florida. American Diabetes Association. Diabetes Care. 1998; 21 (9): 1551-9.
2. Heller GV, Fossati AT. Detection of coronary artery disease in womem. In: Zaret B, Beller G. (eds.). Nuclear cardiology: state of the art and future directions. 2nd ed. Philadelphia: Mosby; 2005.
3. Mieres JH, Shaw LJ, Hendel RC, Miller DD, Bonow RO, Berman DS, et al. American Society of Nuclear Cardiology consensus statement: Task Force on Women and Coronary Artery Disease--the role of myocardial perfusion imaging in the clinical evaluation of coronary artery disease in women [correction]. J Nucl Cardiol. 2003; 10 (1): 95-101.
4. Shaw LJ, Miller DD, Romeis JC, Kargl D, Younis LT, Chaitman BR. Gender differences in the noninvasive evaluation and management of patients with suspected coronary artery disease. Ann Intern Med. 1994; 120 (7): 559-66.
5. Bruce RA. Exercise testing of patients with coronary heart disease: principles and normal standards for evaluation. Ann Clin Res. 1971; 3 (6): 323-32.
6. Azarbal B, Hayes SW, Lewin HC, Hachamovitch R, Cohen I, Berman DS. The incremental prognostic value of percentage of heart rate reserve achieved over myocardial perfusion single-photon emission computed tomography in the prediction of cardiac death and all-cause mortality: superiority over 85% of maximal age-predicted heart rate. J Am Coll Cardiol. 2004; 44 (2): 423-30.
7. Germano G, Erel J, Lewin H, Kavanagh PB, Berman DS. Automatic quantitation of regional myocardial wall motion and thickening from gated technetium-99m sestamibi myocardial perfusion single-photon emission computed tomography. J Am Coll Cardiol. 1997; 30 (5): 1360-7.
8. Gibbons R, Balady GJ, Beasley JW, Bricker JT, Duvernoy WF, Froelicher VF, et al. ACC/AHA Guidelines for exercise testing: (Committee on Exercise Testing). Circulation. 1997; 96: 345-54.
9. Nichols K, Santana CA, Folks R, Krawczynska E, Cooke CD, Faber TL, et al. Comparison between ECTb and QGS for assessment of left ventricular function from gated myocardial perfusion SPECT. J Nucl Cardiol. 2002; 9 (3) :285-9
10. Kinoshita N, Sugihara H, Adachi Y, Nakamura T, Azuma A, Kohno Y, et al. Assessment of transient left ventricular dilatation on rest and exercise on Tc-99m tetrofosmin myocardial SPECT. Clin Nucl Med. 2002; 27 (1): 34-9.
11. Weiss AT, Berman DS, Lew AS, Nielsen J, Potkin B, Swan HJ, et al. Transient ischemic dilation of the left ventricle on stress thallium-201 scintigraphy: a marker of severe and extensive coronary artery disease. J Am Coll Cardiol. 1987; 9 (4): 752-9.
12. Hansen CL, Cen P, Sanchez B, Robinson R. Comparison of pulmonary uptake with transient cavity dilation after dipyridamole Tl-201 perfusion imaging. J Nucl Cardiol. 2002; 9 (1): 47-51.
13 Henzlova MJ, Croft LB, Diamond JA. Effect of hormone replacement therapy on the electrocardiographic response to exercise. J Nucl Cardiol. 2002; 9 (4): 385-7.
14. Klainman E, Kusniec J, Stern J, Fink G, Farbstein H. Contribution of cardiopulmonary indices in the assessment of patients with silent and symptomatic ischemia during exercise testing. Int J Cardiol. 1996; 53 (3): 257-63.
15. Meyer K, Samek L, Pinchas A, Baier M, Betz P, Roskamm H. Relationship between ventilatory threshold and onset of ischaemia in ECG during stress testing. Eur Heart J. 1995; 16 (5): 623-30.
16. Klocke FJ, Baird MG, Lorell BH, Bateman TM, Messer JV, Berman DS, et al. ACC/AHA/ASNC guidelines for the clinical use of cardiac radionuclide imaging--executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/ASNC Committee to Revise the 1995 Guidelines for the Clinical Use of Cardiac Radionuclide Imaging). Circulation. 2003; 108 (11): 1404-18.
17. Douglas PS, Ginsburg GS. The evaluation of chest pain in women. N Engl J Med. 1996; 334 (20): 1311-5.
18. Smanio PE, Watson DD, Segalla DL, Vinson EL, Smith WH, Beller GA. Value of gating of technetium-99m sestamibi single-photon emission computed tomographic imaging. J Am Coll Cardiol. 1997; 30 (7): 1687-92.
19. Taillefer R, DePuey EG, Udelson JE, Beller GA, Latour Y, Reeves F. Comparative diagnostic accuracy of Tl-201 and Tc-99m sestamibi SPECT imaging (perfusion and ECG-gated SPECT) in detecting coronary artery disease in women. J Am Coll Cardiol. 1997; 29 (1): 69-77.
20. Verani MS. Pharmacologic stress myocardial perfusion imaging. Curr Probl Cardiol. 1993; 18 (8): 481-525.
21. Lette J, Tatum JL, Fraser S, Miller DD, Waters DD, Heller G, et al. Safety of dipyridamole testing in 73,806 patients: the Multicenter Dipyridamole Safety Study. J Nucl Cardiol. 1995; 2 (1): 3-17.
22. De Lorenzo A, Lima RS, Siqueira-Filho AG, Pantoja MR. Prevalence and prognostic value of perfusion defects detected by stress technetium-99m sestamibi myocardial perfusion single-photon emission computed tomography in asymptomatic patients with diabetes mellitus and no known coronary artery disease. Am J Cardiol. 2002; 90 (8): 827-32.
23. Di Carli MF, Hachamovitch R. Should we screen for occult coronary artery disease among asymptomatic patients with diabetes? J Am Coll Cardiol. 2005; 45 (1): 50-3.
24. Prior JO, Monbaron D, Koehli M, Calcagni ML, Ruiz J, Bischof DA. Prevalence of symptomatic and silent stress-induced perfusion defects in diabetic patients with suspected coronary artery disease referred for myocardial perfusion scintigraphy. Eur J Nucl Med Mol Imaging. 2005; 32 (1): 60-9.
Paola Emanuela Poggio Smanio Rua João Lourenço, 763/94 - Vila Nova Conceição 04508-031 - São Paulo, SP - Brasil E-mail:
Artigo recebido em 18/6/07; revisado recebido em 31/07/07; aceito em 08/08/07.
BACKGROUND: Cardiovascular disease is the leading cause of morbidity and mortality among diabetic individuals. Myocardial ischemia is frequently asymptomatic, thus leading to a late diagnosis and worse prognosis. Diabetic women are known to have a cardiovascular death risk higher than that in men. OBJECTIVE: To assess the prevalence of coronary artery disease (CAD) in asymptomatic diabetic women. To compare the results of exercise test (ET), cardiopulmonary exercise test (CPET), and dipyridamole myocardial perfusion scintigraphy (MPS) with the findings of coronary angiography, (ANGI) in order of identify the most accurate method in the detection of significant CAD. METHODS: A total of 104 diabetic women were assessed with ET, CPET and MPS in the period within two months from the ANGI. MIBI-99mTc scintigraphy was performed using the gated-SPECT technique. Pearson's chi-square, Student's t tests were used for the statistical analysis and also the logistic regression analysis. RESULTS: The prevalence of CAD in the group studied was 32.7%. For the ET, an effective test (p=0.045), the chronotropic incompetence (p=0.031), and the exercise time performed (p=0.022) showed a significant association with DAC. For CPET, peak VO2 and HR achieved were associated with CAD (p=0.004 and p=0.025, respectively). Most of the MPS variables showed a significant association with CAD (p=0.001, for all). CONCLUSION: The results obtained may suggest a high prevalence of CAD in diabetic women. Thus, this population should be investigated from the cardiovascular point of view even without cardiac symptom. Of the noninvasive diagnostic methods used, dipyridamole MPS was the one that showed the highest discrimination power in relation to diabetic women with CAD.
ORIGINAL ARTICLE
Instituto Dante Pazzanese de Cardiologia – Universidade Federal de São Paulo, São Paulo, SP – Brazil
The prevalence of diabetes mellitus (DM) has been growing fast and continuously in the past decades, thus acquiring epidemic proportions1. Paradoxically to the reduction in coronary artery disease (CAD) mortality, diabetic individuals have presented a significant increase in mortality1. Cardiovascular disease is the leading cause of morbidity and mortality, affecting approximately half of the diabetic patients1. In this group of patients, myocardial ischemia is frequently silent, thus leading to a late diagnosis and worse prognosis1.
Since the 1990's, studies2-4 have demonstrated differences in the investigation of CAD according to the patient's gender, and these differences have been attributed to the lower incidence and atypical symptoms among females. However, diabetic women are known to have a cardiovascular death risk 7.5 times higher than non-diabetic women18. Since the presence of cardiovascular events (CE) may be modified by an appropriate screening and adequate intervention and treatment, early diagnosis may be of great benefit.
The objective of the present study is to assess the prevalence of CAD among type-2 diabetic women without cardiovascular symptoms and also to identify the most effective diagnostic method to detect significant CAD in the group mentioned, by comparing the findings of exercise test (ET), cardiopulmonary exercise test (CPET), and dipyridamole (DIPY) myocardial perfusion scintigraphy (MPS) with those of coronary angiography (ANGI).
Initially, 120 type-2 diabetic women aged between 40 and 80 years and diagnosed with DM for at least five years were included. Women with functional limitation or symptoms suggestive of ischemia were excluded. The patients were recruited (after study approval by the institution's ethics and research committee) via telephone call or mail and were included in the study after a baseline interview, clinical evaluation, and signature of the informed consent form. After the interview, 16 patients were excluded (because of symptoms suggestive of ischemia or difficulty to attend follow-up appointments to undergo tests). A total of 104 patients remained in the study and had their glycosylated hemoglobin (Hb1Ac), ET, CPET, MPS, and ANGI performed with a maximum 2-month interval and with no revascularization procedures between the tests. According to previous recommendation, medications that could interfere with the analysis of the diagnostic tests were discontinued. The tests were analyzed by two specialists in each area, and they had no information on the results of the other tests.
The ET was interpreted, according to the presence of electrocardiographic and clinical alterations, as suggestive of ischemia in the presence of ST-segment depression (ST-depression)
The images were processed by the QGS software program7,9. Gated-SPECT provided a reproduction of cardiac wall contractility, end-systolic and diastolic volume indexes (ESV and EDV, respectively), and left ventricular ejection fraction (LVEF) index. Qualitative assessment of the radiopharmaceutical concentration in 17 myocardial segments was used to analyze the perfusion images. According to the homogeneity, or lack of homogeneity, of uptake in the myocardial segments, MPS were considered as: normal, if the indicator concentration was homogeneous in both phases; suggestive of ischemia, if there was a reversible low indicator uptake in one or more myocardial segments after stress; suggestive of fibrosis if there was a fixed low uptake in one or more segments in both phases; and suggestive of ischemia and fibrosis if there was fixed and reversible low uptake in one or more segments in both phases.
The analysis of left ventricular (LV) systolic function was based on information from the gated-SPECT. The variables obtained were: wall contractility (normal or abnormal); systolic wall thickening (normal or abnormal), and dilation in the stress and baseline phases. The presence of LV dilation was analyzed by taking into consideration the ESV and EDV recommended in the literature. In the QGS software program analysis, ESV equal to 84+/- 26 ml and EDV equal to 33+/-17 ml were considered normal9. Also, the presence of cavity dilation in the stress phase was analyzed in comparison with basal one (transient LV dilation), using visual analysis and the TID (transient ischemic dilation) index, with values higher than 1.22 representing significant dilation10,11. Other variables obtained were the presence of abnormal pulmonary radionuclide uptake12; LVEF in the post-DIPY and baseline phases, with values
After the statistical analysis, a 32.7% prevalence of significant CAD was observed among diabetic women with no cardiovascular symptoms. Of the 104 patients studied, 34 had a
The high prevalence of CAD found in the group studied (32.7%) may be explained by the particular characteristic of our Institution. Given that it is one of the cardiology referral centers in the State of Sao Paulo, it serves a large number of patients with multiple risk factors for CAD, thus characterizing a population with a high pretest probability (
According to the consensus published in 1988 by the American Diabetes Association1, the estimated prevalence of CAD for patients with DM is of 55%.
No studies demonstrating the prevalence of CAD, specifically among women with type-2 DM with no cardiovascular symptoms, are available in the literature.
No clinical and epidemiological characteristic showed a significant association with CAD in the group studied using univariate analysis. Using logistic regression analysis, however, abnormal Hb1Ac values and smoking were factors discriminating patients with CAD in this population (
The first challenge in the diagnosis of CAD in diabetic women is to determine the best method for an initial screening.
Traditional resting ECG adds little information in the identification of patients with CAD, since it only rules out or suggests the presence of acute CE or previous myocardial infarction (MI)13. Baseline ECG in women frequently shows ventricular repolarization abnormalities (VRA), thus making it hardly relevant in the diagnosis of CAD13. In premenopause, ECG specificity is known to be even lower because estrogen-related VRA are more frequent, and estrogen is considered responsible for vasoconstriction of the coronary network, thus resulting in the abnormalities mentioned13. VRA are restricted to the ST-segment and the T wave, and are similar to those observed during the use of digitalis, which has molecules structurally similar to those of estrogen13. Of the 104 patients evaluated, 39 (37.5%) presented VRA on baseline ECG.
ET is largely used as the method of choice in the initial investigation of CAD; according to the literature, it has sensitivity (SE) and specificity (SP) values lower in females than in males (approximately 61% and 69%, respectively)5,8. As previously mentioned, we found lower values in the present study.
Diabetic women also frequently have peripheral vascular insufficiency1. In the group studied, 43 patients (42%) had an ineffective ET, which may be one of the reasons for the low SE found. Another limitation was the large number of tests considered "falsely" suggestive of ischemia because of the VRA. Henzlova et al13 compared three groups of women who had undergone MPS after exercise. Although the MPS had been normal, the patients that were on hormone replacement therapy had 39% of ET compatible with ischemia, in comparison with 22% in the premenopausal group, and 17% in the postmenopausal group, both not receiving estrogens. Of the 39 patients (37.5%) who had a baseline ECG with VRA in the group studied, 10 (33.3%) had ET suggestive of ischemia in the absence of significant CAD, which probably made SP low.
Exercise time (Dt) may indirectly reflect the patients' functional capacity5,8. In the group studied, despite the lower limb limitation and low Dt, a significant association was observed between Dt and the presence of CAD (p=0.022). The heart rate achieved was also associated with CAD. In the group of patients with CAD, HR was, on average, equal to 132.45 bpm, and in the patients without CAD, it was equal to 142.07 bpm, with p=0.027 (
HR is known to increase linearly during dynamic exercise according to the increase in the workload and O2 consumption. A higher HR elevation is usually observed among females in relation to males5, and it is known that an inadequate increase in HR may be an important sign of CAD. Azarbal et al6 verified that CI is an important predictor of mortality and should be incorporated in risk stratification algorithms of ET. However, the authors described that MPS was a better predictor of mortality and of CE in relation to the presence of CI. In the present study, a significant association was observed between CI and CAD (p=0.031). Of the 34 patients with significant CAD, 26 (73.5%) showed criteria of CI.
As previously mentioned, CPET may provide valuable information on functional capacity. In this study, we attempted to assess the value of CPET in the diagnosis of CAD in asymptomatic diabetic women. Because of the lower cost in relation to MPS and of the possibility of association of ECG and clinical variables obtained with exercise with functional parameters, CPET is perhaps a noninvasive diagnostic method with a higher cost-effectiveness ratio than ET and DIPY-MPS.
Few studies on the usefulness of CPET in the investigation of CAD have been published14,15. Klainman et al14 demonstrated that ventilatory variables are decreased with exercise when healthy patients are compared to those with CAD. In the present study, peak VO2 and HR achieved with exercise were observed to be associated with the presence of CAD (p=0.004 and p=0.025, respectively). The SE value found for CPET in the diagnosis of CAD was also low: equal to 44.1%. The SP value, in turn, was higher than that of ET (72.8%). Of the 34 patients with CAD on ANGI, only 15 had a CPET suggestive of ischemia. On the other hand, of the 70 patients without significant CAD, 51 had a CPET not suggestive of ischemia at exercise.
The identification of severity and extent of perfusion defects enables the degree of functional consequences of a coronary stenosis to be established. Since not always does an anatomical coronary artery lesion produce ischemic alterations, there are situations in which a joint anatomical and functional investigation would be the most appropriate management.
In addition to the importance in risk stratification and clinical management, MPS is considered an important noninvasive method that adds diagnostic accuracy to ET for both genders16,17. Its high SP is particularly useful to identify women with ET considered "false negative"8. The mean SE and SP values for planar thallium-201-MPS are 84% and 87%, respectively, and for SPECT-MPS, the mean SE increases to 88%, without decreasing SP (87%). With the advances in methodology, there were increases in SE and SP values. Smanio et al found SE and SP values of 95% and 94.1%, respectively, using the gated-SPECT technique18.
To date, a small number of studies on MPS in women is found in the literature. Taillefer et al19 compared thallium-201 to MIBI-99mTc in 115 women and found no significant difference in SE between the two radiopharmaceuticals19. However, there was a great increase in SP with the gated-SPECT technique (92% vs. 67%, respectively).
Previous studies20,21 found similar values for DIPY and adenosine testing in relation to the images obtained after exercise. Verani et al20 confirmed the superiority of DIPY-MPS in relation to submaximal ET in the diagnosis of CAD. Because of the low functional capacity found in the group studied (43% of ineffective ET), an important investigation was made using a method that did not require strenuous effort and had a high diagnostic power, such as the DIPY-MPS.
Few studies20,25 verified the value of MPS in asymptomatic diabetic patients. For this purpose, the multicenter DIAD study (Detection of Ischemia in Asymptomatic Diabetics) enrolled patients from 14 American and Canadian centers26, and found 22% of ischemia in asymptomatic type-2 diabetic subjects. Rajagopalan et al25 found 826 patients (58%) with ischemia among 1427 asymptomatic diabetics. Prior et al24 and Lerner et al27 identified perfusion defects in 37% of the 133 asymptomatic diabetic individuals (of both genders) who underwent MPS. Of these, 30% were reversible, thus suggesting ischemia.
In this study, SE and SP values of MPS in the diagnosis of CAD were high (94% and 97%, respectively), and similar to those described in the literature16. Most of the MPS variables were associated with CAD (
Functional information is also useful in the diagnosis and prognosis of CAD16, 28. In this study, we verified that baseline and post-DIPY LVEF values showed association with CAD (p=0.001 for both). However, the LVEF decrease after DIPY in relation to baseline did not show a significant association with CAD (p=0.651).
Yamagishi et al's study28 demonstrated that the decrease in LVEF values after stress promotes an increase in SE for the detection of multivessel disease. We can try to explain this discrepancy of the literature by the presence of alterations in microcirculation that are common in diabetics and which can cause subendocardial ischemia and drop in LVEF during pharmacological testing in the absence of an anatomically severe obstructive lesion.
Systolic thickening and contractility alterations also showed a significant association with CAD (p = 0.001). The presence of LV dilation and pulmonary MIBI uptake in the stress phase were also verified to be associated with CAD (p = 0.017 and p = 0.003, respectively). In patients with resting or stress-induced LV dysfunction there can be a radiopharmaceutical extravasation into the pulmonary interstitial space due to the increase in end-diastolic pressure12. Additionally, with the prolonged circulatory time, increased radioactivity is observed in the lung fields. Although this phenomenon is more frequent in thallium-201 MPS, it can also be observed with Mibi-99mTc.
Another variable analyzed was transient left ventricular dilation (TID) after dipyridamole infusion. The presence of TID after the stress phase is usually related to LV failure and has a strong association with CAD with stress-induced ischemic dysfunction, which has an unquestionable predictive value in the determination of survival11. Hansen et al12 studied 1129 patients and verified a significant association between the presence of TID after DIPY infusion and CAD12, and attributed this ventricular cavity dilation to a diffuse subendocardial hypoperfusion. In the group studied, a significant association was observed between the presence of TID and CAD (p = 0.001).
When clinical and electrocardiographic variables during DIPY infusion were analyzed, CP and ECG alterations were observed not to have a significant association with CAD (p = 0.166 and 0.053, respectively). ST-depression during DIPY infusion may be secondary to the "flow steal" phenomenon20. However, no association between ST-depression and significant CC on ANGI was observed in this group either (p = 0.467).
The method proved safe and the side effects presented were similar to those described in the literature21.
The qualitative analysis of the perfusion images was a study limitation. The interpretation using quantitative analysis by scores described in the literature is neither routinely used in our service nor in most of the clinical nuclear medicine services. Therefore, it would not make sense to investigate CAD with a methodology that is not actually used in daily clinical practice.
Finally, we understand that this study is of great value in the clinical management of diabetic women. Knowing that the cardiovascular risk is high in patients with DM, and that women have clinical manifestations which are difficult to characterize, preventive cardiovascular assessment is warranted in the group studied. The presence of VRA on baseline ECG leads to lower accuracy of exercise tests in the identification of CAD. In addition to the frequently observed low functional capacity in diabetic women, the value of methods that depend on the performance of strenuous exercise is limited. On the other hand, it is not cost-effective to perform an anatomical investigation in all diabetic women.
As a consequence of the arguments shown, we can presume that performing a safe technique that has a high diagnostic and prognostic power, such as MPS associated with pharmacological testing with dipyridamole, may be the method of choice in the initial assessment of diabetic women, even in the absence of cardiovascular symptoms.
As a primary conclusion, the results obtained showed that the prevalence of CAD in diabetic women with no cardiovascular symptoms is high (32.7%), and this population should be investigated from the cardiovascular point of view. Also, we conclude that the noninvasive diagnostic method that showed the highest power to discriminate women with CAD was DIPY-MPS; thus, this should be the gold standard method for this group of patients.
No potential conflict of interest relevant to this article was reported.
There were no external funding sources for this study.
This article is part of the thesis of doctoral submitted by Paola Emmanuela Poggio Smanio, from Universidade Federal de São Paulo-UNIFESP
1. Consensus development conference on the diagnosis of coronary heart disease in people with diabetes: 10-11 February 1998, Miami, Florida. American Diabetes Association. Diabetes Care. 1998; 21 (9): 1551-9.
2. Heller GV, Fossati AT. Detection of coronary artery disease in womem. In: Zaret B, Beller G. (eds.). Nuclear cardiology: state of the art and future directions. 2nd ed. Philadelphia: Mosby; 2005.
3. Mieres JH, Shaw LJ, Hendel RC, Miller DD, Bonow RO, Berman DS, et al. American Society of Nuclear Cardiology consensus statement: Task Force on Women and Coronary Artery Disease--the role of myocardial perfusion imaging in the clinical evaluation of coronary artery disease in women [correction]. J Nucl Cardiol. 2003; 10 (1): 95-101.
4. Shaw LJ, Miller DD, Romeis JC, Kargl D, Younis LT, Chaitman BR. Gender differences in the noninvasive evaluation and management of patients with suspected coronary artery disease. Ann Intern Med. 1994; 120 (7): 559-66.
5. Bruce RA. Exercise testing of patients with coronary heart disease: principles and normal standards for evaluation. Ann Clin Res. 1971; 3 (6): 323-32.
6. Azarbal B, Hayes SW, Lewin HC, Hachamovitch R, Cohen I, Berman DS. The incremental prognostic value of percentage of heart rate reserve achieved over myocardial perfusion single-photon emission computed tomography in the prediction of cardiac death and all-cause mortality: superiority over 85% of maximal age-predicted heart rate. J Am Coll Cardiol. 2004; 44 (2): 423-30.
7. Germano G, Erel J, Lewin H, Kavanagh PB, Berman DS. Automatic quantitation of regional myocardial wall motion and thickening from gated technetium-99m sestamibi myocardial perfusion single-photon emission computed tomography. J Am Coll Cardiol. 1997; 30 (5): 1360-7.
8. Gibbons R, Balady GJ, Beasley JW, Bricker JT, Duvernoy WF, Froelicher VF, et al. ACC/AHA Guidelines for exercise testing: (Committee on Exercise Testing). Circulation. 1997; 96: 345-54.
9. Nichols K, Santana CA, Folks R, Krawczynska E, Cooke CD, Faber TL, et al. Comparison between ECTb and QGS for assessment of left ventricular function from gated myocardial perfusion SPECT. J Nucl Cardiol. 2002; 9 (3) :285-9
10. Kinoshita N, Sugihara H, Adachi Y, Nakamura T, Azuma A, Kohno Y, et al. Assessment of transient left ventricular dilatation on rest and exercise on Tc-99m tetrofosmin myocardial SPECT. Clin Nucl Med. 2002; 27 (1): 34-9.
11. Weiss AT, Berman DS, Lew AS, Nielsen J, Potkin B, Swan HJ, et al. Transient ischemic dilation of the left ventricle on stress thallium-201 scintigraphy: a marker of severe and extensive coronary artery disease. J Am Coll Cardiol. 1987; 9 (4): 752-9.
12. Hansen CL, Cen P, Sanchez B, Robinson R. Comparison of pulmonary uptake with transient cavity dilation after dipyridamole Tl-201 perfusion imaging. J Nucl Cardiol. 2002; 9 (1): 47-51.
13 Henzlova MJ, Croft LB, Diamond JA. Effect of hormone replacement therapy on the electrocardiographic response to exercise. J Nucl Cardiol. 2002; 9 (4): 385-7.
14. Klainman E, Kusniec J, Stern J, Fink G, Farbstein H. Contribution of cardiopulmonary indices in the assessment of patients with silent and symptomatic ischemia during exercise testing. Int J Cardiol. 1996; 53 (3): 257-63.
15. Meyer K, Samek L, Pinchas A, Baier M, Betz P, Roskamm H. Relationship between ventilatory threshold and onset of ischaemia in ECG during stress testing. Eur Heart J. 1995; 16 (5): 623-30.
16. Klocke FJ, Baird MG, Lorell BH, Bateman TM, Messer JV, Berman DS, et al. ACC/AHA/ASNC guidelines for the clinical use of cardiac radionuclide imaging--executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/ASNC Committee to Revise the 1995 Guidelines for the Clinical Use of Cardiac Radionuclide Imaging). Circulation. 2003; 108 (11): 1404-18.
17. Douglas PS, Ginsburg GS. The evaluation of chest pain in women. N Engl J Med. 1996; 334 (20): 1311-5.
18. Smanio PE, Watson DD, Segalla DL, Vinson EL, Smith WH, Beller GA. Value of gating of technetium-99m sestamibi single-photon emission computed tomographic imaging. J Am Coll Cardiol. 1997; 30 (7): 1687-92.
19. Taillefer R, DePuey EG, Udelson JE, Beller GA, Latour Y, Reeves F. Comparative diagnostic accuracy of Tl-201 and Tc-99m sestamibi SPECT imaging (perfusion and ECG-gated SPECT) in detecting coronary artery disease in women. J Am Coll Cardiol. 1997; 29 (1): 69-77.
20. Verani MS. Pharmacologic stress myocardial perfusion imaging. Curr Probl Cardiol. 1993; 18 (8): 481-525.
21. Lette J, Tatum JL, Fraser S, Miller DD, Waters DD, Heller G, et al. Safety of dipyridamole testing in 73,806 patients: the Multicenter Dipyridamole Safety Study. J Nucl Cardiol. 1995; 2 (1): 3-17.
22. De Lorenzo A, Lima RS, Siqueira-Filho AG, Pantoja MR. Prevalence and prognostic value of perfusion defects detected by stress technetium-99m sestamibi myocardial perfusion single-photon emission computed tomography in asymptomatic patients with diabetes mellitus and no known coronary artery disease. Am J Cardiol. 2002; 90 (8): 827-32.
23. Di Carli MF, Hachamovitch R. Should we screen for occult coronary artery disease among asymptomatic patients with diabetes? J Am Coll Cardiol. 2005; 45 (1): 50-3.
24. Prior JO, Monbaron D, Koehli M, Calcagni ML, Ruiz J, Bischof DA. Prevalence of symptomatic and silent stress-induced perfusion defects in diabetic patients with suspected coronary artery disease referred for myocardial perfusion scintigraphy. Eur J Nucl Med Mol Imaging. 2005; 32 (1): 60-9.
Paola Emanuela Poggio Smanio Rua João Lourenço, 763/94 – Vila Nova Conceição 04508-031, São Paulo, SP – Brazil E-mail:
Manuscript received June 18, 2007; revised manuscript received July 31, 2007; accepted August 8, 2007.