Abstract
Purpose:
To evaluate the correlation of tear osmolarity with other dry eye parameters in repeated testing in a small cohort of cGVHD subjects with ocular rosacea and severe dry eye disease (DED).
Methods:
Sixteen eyes in 8 subjects with cGVHD, ocular rosacea, and severe DED after allogeneic hematologic stem cell transplantation were evaluated at 9 time points over a 12-month period during a prospective trial of efficacy of IPL and MGX. Among other variables, OSDI, meibomian glands evaluation (MGE), tear break up time (TBUT), and tear osmolarity for each eye were recorded. SAS software was used to perform Pearson correlation coefficients and paired t-test statistical analysis.
Results:
62.5% of subjects were female. Average age was 50 (range 36-64) years. The mean baseline OSDI score was 51.7 (scale 0-100). The mean baseline tear osmolarity was 316 (range 287-365). 73% of eyes had abnormal baseline tear osmolarity (≥308 mOsm/L). Tear osmolarity did not significantly correlate with OSDI, Schirmer’s, TBUT, or MGE. Unexpectedly, the right eyes’ tear osmolarity values weakly positively correlated with Schirmer’s testing (r=0.24, p=0.065). Though the OSDI scores improved 35% with IPL/MGX treatment by 5 months and sustained to 12 months (p<0.001, paired t-test), no improvement in tear osmolarity was found at 5 and 12 months.
Conclusions:
There is considerable controversy about the utility of tear osmolarity in the treatment of DED. Previous researchers have recommended selecting only the worse eye tear osmolarity value as the study eye. Unlike other milder forms of DED, our clinical experience has shown asymmetric disease severity in cGVHD patients, so each eye was evaluated independently at each time interval and recorded. We did not find expected correlation between tear osmolarity and OSDI score. There was no significant correlation between tear osmolarity and other diagnostic tests. Tear osmolarity, OSDI, Schirmer’s, TBUT, and MGE were not interchangeable in the evaluation of DED in our study population. The clinical utility of monitoring severe DED in cGVHD with tear osmolarity was not apparent with the results of this limited study.