Can Light Therapy Before Cataract Surgery Prevent Dry Eye?

Dry eye after cataract surgery is common enough to deserve attention before the first incision. A 2025 double-masked randomized trial tested whether low-level light therapy (LLLT) could blunt that postoperative decline rather than chase symptoms afterward. The answer was encouraging, but far from settled: patients receiving perioperative LLLT had better symptom and ocular-surface results at one month than a sham group, in a study with important gaps in reporting and follow-up.

A preventive idea with one important wrinkle

Phacoemulsification can disrupt the ocular surface through corneal nerve injury, inflammation, and meibomian-gland disturbance. Those changes can trigger or worsen dry eye, especially in people who already have a fragile tear film.

In the 2025 trial by Timofte-Zorila et al., 89 people undergoing cataract surgery were randomly assigned to active treatment or a sham procedure in a prospective, double-masked design. The active group received two 15-minute sessions with the eye-light® device at 633 ± 10 nm: one seven days before surgery and one seven days after. The sham device delivered less than 30% of therapeutic power.

At one month, the groups had separated on the Ocular Surface Disease Index (OSDI), tear osmolarity, and Oxford ocular-surface staining. The sham group showed postoperative rises in tear osmolarity and surface damage; the LLLT group did not. The reported effect size for OSDI was large, with Cohen's d = 0.93 and post-hoc power of 99%. The Schirmer test, which measures tear production, did not change significantly in either group.

That pattern fits a possible effect on the evaporative side of dry eye: better tear-film stability and less surface injury without increased aqueous tear secretion. It does not prove that LLLT prevented dry eye in people with previously normal eyes, though. The average baseline OSDI was about 27, a mild-to-moderate symptomatic range, despite exclusions for diagnosed ocular-surface disease or dry eye. Many participants may therefore have entered the study with undiagnosed or subclinical meibomian gland dysfunction.

The trial may have reduced the impact of cataract surgery on an already vulnerable ocular surface. It cannot tell us whether the same treatment prevents dry eye in patients with a genuinely normal tear film.

What the proposed mechanism can—and cannot—explain

Photobiomodulation is thought to act partly through cytochrome c oxidase in mitochondria. Red or near-infrared light may increase ATP production and alter downstream inflammatory signaling. In this setting, the proposed target is the meibomian gland, whose lipid secretion slows tear evaporation.

The cataract trial's outcome pattern is compatible with that explanation, but compatibility is not proof. The investigators did not report direct measures of cytochrome c oxidase activity, ATP, meibography, or lipid-layer thickness. A stable Schirmer result also has more than one interpretation: LLLT may chiefly affect the lipid layer, or it may simply leave the lacrimal-gland and aqueous-deficient components of dry eye untouched.

That distinction matters because dry eye has several causes. In a controlled study of severe dry eye in perimenopausal women, Li et al. reported benefit from estrogen replacement combined with autologous serum therapy. That result does not refute LLLT, but it shows why a meibomian-gland-focused intervention cannot cover every hormonal, immune, or aqueous-deficient form of dry eye.

Other surgical studies support plausibility, not an eye-care standard

Researchers have tested preemptive photobiomodulation in other surgical settings. A randomized split-mouth trial by Cetira Filho et al. paired preoperative photobiomodulation with nimesulide before third-molar surgery and found less postoperative pain, lower oxidative-stress markers, and less quality-of-life impairment. Because every participant also received an NSAID, the study cannot isolate the size of the light-treatment effect.

A separate randomized trial by Karşıcı et al. tested preoperative LLLT alone before mandibular third-molar extraction, examining postoperative pain, swelling, and trismus. It provides a cleaner precedent for treating tissue before a planned surgical insult. Dental tissue, however, is not a meibomian gland, and light penetration, anatomy, and inflammatory biology differ.

Postoperative evidence points in the same broad biological direction without answering the cataract question. After total knee arthroplasty, a three-arm randomized trial by Bahrami et al. found that LLLT improved pain and function and reduced IL-1β, IL-6, and TNF-α. Those findings support anti-inflammatory plausibility after surgical trauma. They do not establish the right dose, target, or schedule around cataract surgery.

LLLT outcomes also vary by clinical context and protocol. A randomized dental study comparing cryotherapy with LLLT for pain after treatment of symptomatic apical periodontitis, reported by KungumaSupraja et al., illustrates that results do not transfer automatically even across dental procedures. Reviews in other conditions likewise report substantial variation in wavelength, energy density, treatment frequency, and outcomes, including a systematic review of photobiomodulation for Bell's palsy.

The trial's limits are clinical limits

The cataract study has real design strengths: randomization, masking, and a sham comparator. But its registration was retrospective. The trial was registered in July 2025 after enrollment from January through March 2025. Retrospective registration makes it harder to verify that the primary outcome and analysis plan were fixed before results became known.

The reported sham also delivered less than 30% of therapeutic power rather than zero power. A partial-dose comparison may reduce the measured contrast between groups, but it does not provide a pure biological null. More practically, the paper does not report power density or total fluence in the available text. A wavelength and session length are not enough to reproduce a photobiomodulation protocol, and they offer no basis for assuming that a consumer light device delivers the same treatment.

Follow-up ended at one month. That window cannot distinguish a durable change in gland function from a temporary reduction in postoperative inflammation. It also cannot show whether the sham group would catch up by three or six months. Full numerical between-group results, including mean differences, confidence intervals, and raw p-values, are unavailable in the published text accessible here, so readers cannot independently assess the precision or clinical size of each endpoint.

What to do before cataract surgery

For someone with known or suspected meibomian gland dysfunction, perioperative LLLT is reasonable to discuss with an ophthalmologist as an adjunct—not a replacement for a dry-eye evaluation and treatment plan. The one randomized cataract trial produced a large symptom effect and favorable changes in tear osmolarity and staining, but it has not yet been replicated and does not establish long-term benefit.

• Ask for a preoperative ocular-surface assessment, especially if you have burning, fluctuating vision, contact-lens intolerance, or an elevated OSDI score.
• Ask whether your dry eye appears mainly evaporative, aqueous-deficient, inflammatory, hormonal, or mixed. Treatment should match the subtype.
• If LLLT is offered, ask for the exact device, wavelength, power density, fluence, treatment timing, and expected out-of-pocket cost.
• Do not assume an at-home red-light device reproduces the clinical protocol.

The evidence supports a specific, modest conclusion: perioperative LLLT may protect a symptomatic or subclinically compromised ocular surface during cataract surgery. It does not yet show that light therapy prevents dry eye in every cataract patient, or that its benefits last beyond the first postoperative month.