Design: Prospective interventional diagnostic study, single centre
Population: 261 margins from 42 non-melanoma skin cancers in 39 patients undergoing Mohs surgery
Primary outcome: Agreement with conventional histopathology
Effect: Accuracy 96.4%; inter-reader κ 0.863 for margins

An Italian prospective study examined 261 margins from 42 non-melanoma skin cancers in Mohs surgery using ex vivo fluorescence confocal microscopy, which produces digitally stained images resembling haematoxylin and eosin without freezing or cutting sections.

A dermatologist and a pathologist, each trained in the method, agreed almost perfectly on margins (κ 0.86) and tumour type (κ 0.88). Against conventional histology, overall accuracy was 96.4%, with one false-negative margin.

The technique preserves tissue for later histology and may speed turnaround, but time savings were not measured. Equipment cost will limit uptake in most Indian centres for now.

Recognise confocal microscopy images as a validated intraoperative margin method in trained hands.
Keep conventional histology as the reference while local validation is under way.
Expect a learning curve; readers here were specifically trained.
Note that tissue remains available for permanent sections after imaging.

Why it matters

It could remove the cryostat step from intraoperative skin margin assessment.

The statistics, in plain English

A kappa above 0.8 is conventionally 'almost perfect' agreement beyond chance. One false-negative among 261 margins is reassuring, but the study is small.