The incremental diagnostic value and clinical utility of tear MMP-9 for dry eye disease: A multicentre analysis and development of a simplified screening model.

OBJECTIVE

To evaluate the incremental diagnostic value of tear matrix metalloproteinase-9 (MMP-9) in dry eye disease (DED) and validate a simplified diagnostic model suitable for primary care.

METHODS

This multicentre study analysed 580 participants (294 DED, 286 controls) from three Centres. Diagnostic performance was evaluated using Generalised Linear Mixed-Effects Models (GLMMs) to adjust for Centre heterogeneity. Three models were compared: (1) Traditional (Ocular Surface Disease Index [OSDI], non-invasive tear breakup time [NIBUT], Schirmer I test, and corneal fluorescein staining [CFS]), (2) Trad + MMP-9, and (3) Simplified (OSDI, Schirmer I test, and MMP-9). Performance was assessed via area under the receiver operating characteristic curve (AUC), Net Reclassification Improvement (NRI), Decision Curve Analysis (DCA), and Leave-One-Centre-Out Cross-Validation (LOCO-CV).

RESULTS

The Trad + MMP-9 model achieved the highest discrimination (AUC 0.982) and calibration. Adding MMP-9 yielded significant risk reclassification improvement (Continuous NRI 0.495, p<0.001) over the Traditional model. A Simplified model (OSDI, Schirmer I test, MMP-9) achieved an AUC of 0.969, statistically comparable to the Traditional model, and demonstrated robust external validity in LOCO-CV. Tear MMP-9 levels correlated strongly with symptom burden (OSDI) independent of clinical signs.

CONCLUSION

Tear MMP-9 provides significant incremental diagnostic value, enabling a simplified, high-accuracy diagnostic strategy that omits slit-lamp examination, potentially expanding DED screening capabilities.