[Specimen misidentification in pathology laboratory: trends and measures]

Specimen misidentification in pathology laboratories may have serious consequences. Reports on the frequency of errors in pathology laboratories in Japan are rare. We reviewed near-miss and incident reports over 7 years in our laboratory, extracted those associated with misidentification, analyzed annual changes in numbers and content, and discussed the problems faced and measures taken to prevent misidentification. Of 113,447 pathological cases, 88 (0.078%) reports were associated with misidentification. Of these 88 misidentification cases, 19% occurred before and during accessioning, 16% during dissecting and sectioning, 30% during embedding, 13% during tissue cutting and slide mounting, 19% during slide submitting, and 3% during diagnosis. Two cases (2.3%) of misidentification were detected after diagnosis; however, misidentification did not appear to cause adverse effects in any patient. The frequency of events is similar to that reported in the literature; specimen misidentification was noted in 0.1-0.2% of cases in a modern pathology laboratory. Two-thirds of misidentification events occurred associated with gross specimens, similar to findings in other studies. With the introduction of new technologies that minimize the possibility of human errors (e.g., barcode reading, digital imaging of every specimen, and installation of a glass slide printer), education on medical safety, and the use of multiple safety nets (e.g., diagnosis cancelling and slide checking), errors have decreased, but have not been eliminated completely. Recording errors and reporting them to the hospital and social community, and maintaining a sustainable quality improvement system is very important to reduce errors in pathology.