There is a lack of systematic investigations comparing optical coherence tomography (OCT) with histology. OCT assessments were performed on the upper back of 16 healthy subjects. Epidermis thickness (ET) was assessed using three methods: first, peak-to-valley analysis of the A-scan (ET-OCT-V); second, manual measurements in the OCT images (ET-OCT-M); third, light microscopic determination using routine histology (ET-Histo). The relationship between the different methods was assessed by means of the Pearson correlation procedure and Bland and Altman plots. We observed a strong correlation between ET-Histo (79.4+/-21.9 microm) and ET-OCT-V (79.2+/-15.5 microm, r=0.77) and ET-OCT-M (82.9+/-15.8 microm, r=0.75), respectively. Bland and Altman plots revealed a bias of -0.19 microm (95% limits of agreement: -27.94 microm to 27.56 microm) for ET-OCT-V versus ET-Histo and a bias of 3.44 microm (95% limits of agreement: -24.9 microm to 31.78 microm) for ET-OCT-M versus ET-Histo. Despite the strong correlation and low bias observed, the 95% limits of agreement demonstrated an unsatisfactory numerical agreement between the two OCT methods and routine histology indicating that these methods cannot be employed interchangeably. Regarding practicability, precision, and indication spectrum, ET-OCT-V and ET-OCT-M are of different clinical value.