The diagnostic yield of histologic examination of seborrheic keratoses

To examine the diagnostic yield in submitting clinically diagnosed seborrheic keratoses for routine microscopic examination.

Retrospective examination of preoperative and postoperative diagnoses based on information provided by the clinician on the laboratory worksheet and the subsequent histopathologic diagnosis.

A regional nonhospital-based dermatopathology laboratory with specimens submitted by physicians (dermatologists and nondermatologists) practicing in a 4-state midwestern region of the United States.

A total of 5592 cutaneous pathology reports were reviewed. Specimens submitted with a preoperative clinical diagnosis of seborrheic keratosis, with or without a modifier, were examined. A comparison group with the clinical diagnosis of melanocytic nevus was reviewed.

Preoperative clinical diagnoses were compared with the microscopic diagnoses.

Of 577 specimens clinically diagnosed and submitted as seborrheic keratoses, 37 (6.4%) were histologically diagnosed as malignant tumors. The rate of malignant tumors increased when clinical information suggested findings beyond the classic clinical presentation, such as irritation, or when a malignant tumor was considered in the differential diagnosis. Two lesions that histologically proved to be melanomas were in this group. Comparison of the seborrheic keratosis group with the nevus group showed that seborrheic keratoses were more likely to be malignant tumors than were melanocytic nevi. Clinically diagnosed seborrheic keratoses submitted by dermatologists were more likely than clinically diagnosed melanocytic nevi to be melanomas.

Our data suggest that there were differences in the rate of malignant tumors between dermatologists and nondermatologists and that clinically diagnosed, surgically removed seborrheic keratoses are more likely than clinically diagnosed, surgically removed melanocytic nevi to be malignant tumors.