Efficacy and safety of intense pulsed light of upper and lower eyelids in Meibomian gland dysfunction: A prospective multicentric study

Abstract

Purpose To demonstrate that intense pulsed light therapy (IPL) of the upper and lower eyelids with meibomian gland expression (MGX) is effective in improving dry eye disease due to meibomian gland dysfunction (MGD). Methods Patients with ocular discomfort (Ocular Surface Disease Index -OSDI- above 13) and signs of MGD were recruited. All patients underwent OSDI, visual acuity (VA), intraocular pressure, Schirmer test, meibography, non-invasive tear breakup time (NITBUT), slit-lamp examination (corneal and conjunctival staining, hyperemia, gland expressibility, and meibum quality), tear osmolarity and lipid layer thickness. IPL was performed with Optima IPL (Lumenis Ltd.) following a standardized protocol on upper and lower eyelids of both eyes, with inferior eyelid MGX. Patients received four sessions separated by two weeks each. Four weeks after, examinations were repeated. Results 160 patients (320 eyes) were included, of which 108 (67.5%) were women and mean age was 59.2 ± 15.08 (range 20–89). After four sessions, VA, OSDI, tear osmolarity, lipid layer thickness, NITBUT, hyperemia, corneal and conjunctival staining, gland expressibility, meibum quality, inferior eyelid Meiboscore and Schirmer test improved (all, p < 0.027). Changes in OSDI, initial and average NITBUT increased with dry eye disease severity (according to OSDI). Increased pre-treatment OSDI, hyperemia, corneal and conjunctival staining and Schirmer test were associated with an improvement in OSDI (all, p < 0.040). No adverse events were noted. Conclusions The combination of IPL on upper and lower eyelids with MGX is safe and effective for the treatment of MGD. Patients with severe dry eye disease present greater improvements.

Intense Pulsed Light Therapy for Dry Eye Disease: Analyzing Temporal Changes in Tear Film Stability and Ocular Surface between IPL Sessions

“…To recognize a statistically significant difference greater than or equal to 0.05, 140 eyes were necessary. The standard deviation was assumed to be 3.8 score points (based on Benitez-del-Castillo et al [ 38 ]) with an anticipated a drop-out rate of 10%. Mean values and standard deviations (SD) were used to summarize continuous data, along with ranges, whereas frequencies (n) and percentages (%) were used for ordinal categorical data.…”