Densitometry of meibometer samples from normal and dry eye subjects

Abstract

Abstract Purpose: To compare the original and modified techniques of meibometry, and scanning/densitometry of lipid blots, on normal and dry‐eye subjects. Methods: A hand‐held loop of plastic meibometry tape was pressed on the lower lid margin (10sec contact at 15g load), or fixed in the prism‐holder of the Goldmann tonometer arm (3sec contact at zero load) to obtain meibomian blots from right and left eyes of the same subject. The tape blot was scanned with the IRIS‐Pen hand‐held scanner to obtain digital images of the blot area and background (unblotted) tape. Mean pixel density of the blot area was found with the NIH ImageJ software and mean background density subtracted. Some blots were measured with both the original Meibometer instrument and by scanning. Results: Comparison of blot densities by the Meibometer and scanning/densitometry on 25 normals showed no significant difference, irrespective of background levels, making the Meibometer unnecessary. A good correlation was also seen between the original and the modified application techniques despite the lower pick‐up of the latter. This is also preferred where patients cannot tolerate the 10sec application without blinking, as in many dry‐eye cases. Mean pixel density was lower in 14 meibomian gland dysfunction patients (6 with, 8 without aqueous tear deficiency) than in normals, although the difference was not significant; readings were not always reproducible due to scanning errors, possibly indicating local tape saturation or composition‐dependent effects. Conclusions: IRIS Pen densitometry gives comparable results to the original Meibometer, so the Pen and a laptop computer can be used in the clinic or in field studies. More work is needed on scanning reproducibility and the effect of other environmental factors.