Influence of baseline characteristics on subjective improvement of dry eye after intense pulsed light therapy

To identify baseline clinical signs and symptoms associated with response to intense pulsed light (IPL) combined with meibomian gland expression in dry eye disease (DED), and to develop machine learning (ML) models for individualized outcome prediction. This retrospective study analyzed 100 eyes fro...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Contact lens & anterior eye s. 102514
Hlavní autoři: Carrillo-Pulido, Miriam, Ortiz-Peregrina, Sonia, López Pérez, María Dolores, Cano-Ortiz, Antonio, González-Cruces, Timoteo
Médium: Journal Article
Jazyk:angličtina
Vydáno: England 07.10.2025
Témata:
Intense pulsed light therapy
Dry eye disease
Meibomian gland dysfunction
Predictive algorithms
Machine learning
ISSN:1367-0484, 1476-5411, 1476-5411
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:To identify baseline clinical signs and symptoms associated with response to intense pulsed light (IPL) combined with meibomian gland expression in dry eye disease (DED), and to develop machine learning (ML) models for individualized outcome prediction. This retrospective study analyzed 100 eyes from 100 DED patients (aged 58.6 ± 13.4 years) treated with IPL and meibomian gland expression. Baseline parameters assessed with the Antares system included meibomian gland loss (MGL), tear meniscus height (TMH), non-invasive tear break-up time (NIBUT), conjunctival hyperemia, and Ocular Surface Disease Index (OSDI). Patients were stratified by change in OSDI after treatment (ΔOSDI): Class 1 (no improvement), Class 2 (mild improvement), and Class 3 (clear improvement). Several ML models were trained to predict ΔOSDI from baseline parameters. IPL significantly improved both symptoms and signs. OSDI decreased from 44.65 ± 18.3 to 28.47 ± 19.3 (p < 0.001), NIBUT increased from 4.5 ± 3.2 to 7.5 ± 6.5 s (p < 0.001), and TMH and conjunctival hyperemia also improved (p < 0.001), while MGL and BCVA remained stable. Greater improvement was observed in patients with higher baseline OSDI (p = 0.001). The XGBoost algorithm achieved the highest predictive performance (AUC-ROC = 0.77), with OSDI and NIBUT as the strongest predictors based on SHAP analysis. IPL combined with meibomian gland expression improves symptoms and signs in DED, particularly in patients with more severe baseline symptoms. Baseline OSDI and NIBUT were the strongest predictors of response. ML models demonstrated moderate accuracy, supporting their potential role in personalized DED treatment strategies.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1367-0484
1476-5411
1476-5411
DOI:10.1016/j.clae.2025.102514