Comparison of tear lipid profile among basal, reflex, and flush tear samples.

Abstract

To determine whether tear collection by flushing the ocular surface with saline (flush tears) or collection by stimulation (reflex tears) can be used as an alternative to basal tear collection for the identification and quantification of lipids in the tear film. Tear samples were collected from 10 participants with no history of ocular surface disease or contact lens wear. Up to 10 μl of basal, reflex, and flush tear samples were collected from each eye using a microcapillary tube on three occasions with the order of methods randomized and allowing at least 24 hours between each collection method. Lipids were quantified from each tear sample using nano-electrospray ionization tandem mass spectrometry. Total lipids significantly differed in their concentration (pmol/μl) and mole % with each collection technique. Cholesterol esters [mean % (SE)] formed the major component of the total lipidome in basal [54.8% (3.1)], reflex [35.7% (6.4)], and flush [33.0% (3.1)] tear samples. However, the mole % of each lipid class substantially varied with each tear collection method. Nonpolar lipids, including cholesterol, wax esters, and triacylglycerols, dominated the tear lipidome in basal [92.8% (1.9)], reflex [71.8% (7.9)], and flush [83.6% (3.8)] tear samples. However, the mole % of phospholipids in reflex [27.5% (8.1)] and flush [15.8% (3.8)] tear samples was higher (p = 0.005) than that in basal tears [5.4% (2.0)]. Flush or reflex tears did not have similar lipid profiles in either concentration or in mole % to basal tears. It is recommended that basal tears are used for tear lipid analysis as the reflex or flush tears contain very low levels of most lipid components.