Findings from a recent case-control study published in BMC Ophthalmology tied antidepressant use to goblet cell loss, conjunctival epithelial damage, and worse dry eye signs, with the pattern holding across every drug class tested.
Give me some background first.
We’ll begin with the mucin layer of the tear film—which is produced by conjunctival goblet cells—and its ability to keep the ocular surface stable and hydrophilic.
When goblet cells are damaged or lost however, mucin supply drops and the tear film destabilizes.
Bringing in antidepressants: Use of these drugs has been linked to ocular dryness through their effect on tear secretion.
But what prior research has yet to explore is the cellular picture—specifically what happens to goblet cells and the conjunctival epithelium.
Now to this new study.
Researchers ran a case-control study to quantify goblet cell density (GCD) and histomorphological changes in the conjunctival epithelium of antidepressant users, then correlated GCD with routine dry eye tests.
The question: Does antidepressant use measurably damage the conjunctival epithelium, and is that damage associated with tear film instability?
Who were the participants?
The study enrolled 142 participants (ages 20-40):
- 72 antidepressant users as cases
- Average age: 30.7; 58.3% male
- 70 age /sex-matched controls
The narrow age band was chosen to rule out age-related dry eye.
Median treatment duration was 12 months (Interquartile Range [IQR] 6 to 16).
And the drug breakdown of these patients?
With depression and anxiety the most common diagnoses, the top three drugs for these patient cases included:
- 75% using selective serotonin reuptake inhibitors (SSRIs)
- 13.9% using tricyclic antidepressants (TCAs)
- 11.1% using serotonin-norepinephrine reuptake inhibitors (SNRIs)
Each participant completed the Ocular Surface Disease Index (OSDI) questionnaire and underwent tear break-up time (TBUT) testing, Schirmer testing, corneal and conjunctival staining, and conjunctival impression cytology.
Next up: the findings.
Cases had sharply lower GCD: 183 cells/mm² versus 366 in controls (P < 0.001).
Tear film function also declined with lower:
- TBUT: 6.7 seconds versus 10.8 (P < 0.001)
- Schirmer wetting: 5.8 mm/5 min versus 7.6 (P = 0.001)
- Nuclear-cytoplasmic (NC) fraction: 0.34 versus 0.41 (P < 0.001)
OSDI scores were roughly twice as high in the antidepressant group: 25.6 versus 12.3 (P < 0.001).
And looking at the odds ratio?
Antidepressant users were far more likely to show corneal staining (odds ratio [OR] 13.0), reduced GCD (OR 9.3), and abnormal TBUT (OR 7.0).
The link: GCD correlated positively with TBUT (r = 0.41) and Schirmer results (r = 0.43), pointing to mucin-producing goblet cell loss as a driver of tear film instability.
Tell me more.
A multiple linear regression model explained about 48% of the variance in GCD (adjusted R² = 0.480, P < 0.001).
Independent predictors: antidepressant use (β = 95.17, P = 0.001), lower TBUT (β = 11.91, P < 0.001), and lower Schirmer scores (β = 20.92, P < 0.001) each independently predicted reduced GCD.
What didn’t predict it: Age, OSDI symptom scores, and both corneal and conjunctival staining fell out of the model, which points to the goblet cell loss being structural rather than symptom or age driven.
Also worth noting: Treatment duration had no correlation with GCD (r = -0.03), so the damage did not simply accumulate with longer use.
Any limitations?
The authors flagged several—starting with the single-center design and convenience sampling raising the risk of selection bias and limiting how far the findings could be generalized.
Others included:
- Lumping SSRIs, SNRIs, and TCAs together introduced pharmacological heterogeneity
- With a small, unevenly distributed sample across drug classes, the study could not confirm whether the effects were drug-specific
- The cornea and conjunctiva were also graded without a standardized system, which may have affected consistency
Expert opinion?
The authors argued that routine ocular surface evaluation should be standard for patients starting antidepressants, with early management through lubricants, mucin-stimulating agents, and coordination between psychiatrists and eyecare providers.
On mechanism: They pointed to anticholinergic activity for TCAs, which blocks the muscarinic signaling goblet cells need, and inflammatory NF-κB and MAPK pathway activation for SSRIs.
Anything else?
One finding that stands out: No statistically significant differences showed up across the SSRI, SNRI, and TCA subgroups, so the ocular risk appears to apply to antidepressants broadly rather than to one class.
- The TCA group did trend worse on several measures—consistent with anticholinergic effects—but the difference was not significant.
Go on …
Also of interest: Conjunctival staining was significantly associated with abnormal GCD grades (P = 0.015), which the authors put forward as a practical biomarker for goblet cell loss.
They flagged the finding as especially relevant for contact lens wearers, where goblet cell function matters more.
Take home.
For patients on antidepressants of any class: The ocular surface takes a measurable hit, with goblet cell density and tear stability affected most.
As such: ECPs may want to consider building routine ocular surface checks into psychiatric care as well as managing early with lubricants and mucin-stimulating agents rather than waiting for symptoms.