[Thermography of the breast. Sensitivity, specificity and reproducibility].

Thermography is a questionable method of diagnosis of breast diseases. In our study, we demonstrate that the low sensitivity and specificity is not related to the experience of the reader. One hundred thermograms were given to five different readers: three of them were thermographists, one was a gynecologist and the last one radiotherapist. Sixteen patients had a breast carcinoma, 18 had benign lesions. Seventy-two patients had no lesion at the time of diagnosis and 4 years after. The false negative and the false positive rates were not different according to the readers. The reproducibility, judged with the Kappa test, was poor. This study, in agreement with most of the previous studies, demonstrates that thermography must not be recommended neither for diagnosis nor for screening of breast diseases.