Accéder directement au contenu Accéder directement à la navigation
Communication dans un congrès

Definition of peripheral arterial disease by non-invasive diagnostic methods for epidemiological studies.

Abstract : In order to establish a registry of patients with peripheral arterial disease (PAD), a precise definition of this disease is mandatory. Arteriography cannot be used in large scala studies. Non-invasive methods as the WHO questionnaire, clinical examination, ankle-arm index (AAI) and Doppler are used in epidemiological studies but there is no consensus since each method is insufficient to take account of the whole range of this disease. The Jack of standard definition makes comparison between studies difficult. The purpose of this study was to elaborate a standard combination of these diagnostic tools to be able to distinguish patients with or without PAD for enrolment in epidemiological surveys. Methods: Two hundred patients addressed to our vascular laboratory for a suspicion of PAD answered to the WHO questionnaire, had clinical and Doppler examination and AAI measurement. The results were compared to the diagnosis of a panel of experienced angiologists, as gold-standard. A Receiver Operating Characteristic (ROC) curve was used to define the lower cut-off point of AAI. An upper cut-off point is generally admitted at 1.5. All variables of the different diagnostic methods were then coded as "normal" or "abnormal". An automatic segmentation by SIPINA software ® based on Shannon's entropy was used to separate patients with and without PAD by the binary variables. By this segmentation an algorithm was elaborated for the diagnosis. With a Monte-Carlo simulation, the reproducibility of the algorithm was studied. Results: The ROC curve showed a lower cut-off point of AAI at 0.85. With only three binary variables, respectively normal or abnormal physical examination, AAI and Doppler pattern, an algorithm was elaborated which had a sensitivity of 95.4%, specificity of 93.5%, and an accuracy of 94.5%. The Monte-Carlo simulation with 1000 replications revealed an accuracy of 91.8%, confirming the good reproducibility of the algorithm. Conclusion: A combination of non-invasive methods was necessary to perform a precise diagnosis of PAD. This algorithm is an interesting toot to standardise the diagnosis of PAD in epidemiological studies
Type de document :
Communication dans un congrès
Liste complète des métadonnées
Contributeur : Elisabeth Grelier <>
Soumis le : jeudi 5 mars 2015 - 11:48:25
Dernière modification le : lundi 12 mars 2018 - 10:26:02


  • HAL Id : hal-01123658, version 1



Victor Aboyans, J. Ferrieres, Philippe Lacroix, Alain Vergnenègre, Pierre-Marie Preux, et al.. Definition of peripheral arterial disease by non-invasive diagnostic methods for epidemiological studies.. Communication orale présentée au "XXème Congrès de la Société Européenne de Cardiologie", Aug 1998, Vienne, Austria. pp.458. ⟨hal-01123658⟩



Consultations de la notice