Correlation of conjunctival impression cytology results with lacrimal function and age

Abstract

Impression specimens were obtained from the bulbar conjunctiva of 146 eyes of 73 healthy volunteers (mean age, 52 years). Goblet cell densities and nuclear-cytoplasmic (N/C) ratio were determined. Break-up time (BUT), Schirmer-I test (S-I), Rose Bengal scores (RBs) were also recorded. Using a simple linear regression analysis, no correlation between the number of Goblet cells and age, BUT and S-I was found. Negative correlation with RBs (P = 0.05) was found. The N/C ratio has a negative correlation (P less than 0.01) with age and RBs and a positive correlation with BUT (P less than 0.01) and S-I (P = 0.03).

Goblet cell density association with tear function and ocular surface physiology

“…In consistent to this study, Yeo et al [7] and Paschides et al [28] also did not find any relation of GCD with phenol red thread test score. It might be because there is little or no direct role of goblet cells on the aqueous portion of the tears.…”

“…However, the correlation was weak between GCD and TBUT. Yeo et al [7] and Paschides et al [28] did not find any correlation of GCD with NIBUT and TBUT. Yeo et al [7] suggested that, TBUT depends not only up on MUC5AC which is produced by goblet cells but also with other mucin produced by other conjunctival cells.…”

The Correlation of Routine Tear Function Tests and Conjunctival Impression Cytology in Dry Eye Syndrome

“…Paschides et al [24] also found a similar correlation using a simple linear regression analysis to compare the results. These authors reported that 24.1% patients with dry eye had a normal RBS score and 20.6% had normal cytology.…”

Sampling Area Selection for the Assessment of Goblet Cell Density From Conjunctival Impression Cytology Specimens

“…However, on some CIC specimens, a calibrated grid was used at ·400 to therefore providing a sampling area of just 0.008 mm 2 . 28 Others have reported assessing the goblet cell count from a ·200 microscope field without specifically stating what the sampling area size was, 29,30 whereas others reported using sampling area consistent with the use of a ·20 objective lens, that is, of 300 · 300 mm 2 (i.e., 0.09 mm 2 ) 31 or 260 · 260 mm 2 (i.e., 0.068 mm 2 ). [23][24][25] Others report just 0.01 mm 2 for the sampling area, 26 that is, similar to that expected from the use of a ·400 imaging grid.…”