Comparability among histopathologists in the diagnosis and classification of lesions suspected of leprosy in Malawi.

Identical slides from 200 biopsies obtained from individuals suspected of having leprosy during the course of an epidemiological survey in Northern Malawi were examined sequentially and independently by three histopathologists, using a standard protocol to report their findings. Their results are compared among themselves and with a standardized clinical assessment of each subject. There was more agreement among the histopathologists as to classification of leprosy cases than there was on the diagnosis of leprosy. The proportion of biopsies considered to show definite evidence of leprosy varied from 39% to 58% among the histopathologists. An appreciable additional proportion of biopsies (11.5% to 38.5% for the three histopathologists) was considered to show evidence suggestive but not pathognomonic of leprosy. Although there was, in general, good agreement on classification, the proportion of biopsies considered to show evidence of indeterminate leprosy varied from 1.5% to 21.5% among the three histopathologists. This suggests that some of the reported differences in the prevalence and proportion of indeterminate leprosy in different populations is due to terminology alone. A follow-up meeting of the study participants revealed that many of the differences in diagnosis certainty were due to difficulties in interpreting evidence of nerve involvement. It is recommended that greater attention be paid to the difficulties of diagnosing leprosy on histopathological as well as clinical grounds. A revised standard protocol for reporting histopathological evidence of leprosy is presented.