Findings from a recent study published in the American Journal of Ophthalmology compared the incidence and timing of postoperative complications after laser-assisted in situ keratomileusis (LASIK) and photorefractive keratectomy (PRK).
The data point to meaningful differences in when—and how often—patients run into trouble after each procedure.
Give me some background first.
LASIK and PRK are two of the most common elective procedures in ophthalmology, with comparable visual outcomes. The difference is in how the cornea heals afterward, which gives each procedure its own complication profile.
Some context: The Academy's Refractive Surgery Preferred Practice Pattern (PPP) flags dry eye disease (DED), visual disturbances, and ocular pain as the complications seen most often after both procedures.
So what’s the issue?
How that risk actually plays out over time, in routine clinical practice, has been poorly characterized.
Worth noting: Baseline DED is already common in refractive candidates, with reported prevalence ranging from 25.9% to 53.3% depending on the population (as documented in candidate cohorts).
Now, talk about the study.
Researchers ran a retrospective registry-based cohort study using the IRIS Registry (Intelligent Research in Sight), pulling refractive surgeries performed between January 2013 and March 2024.
They identified complications from diagnostic codes and tracked incidence across three postoperative windows: 1 day to 1 month, 1 to 3 months, and 3 to 12 months.
And how was this healing defined?
An eye counted as having postoperative DED only when all criteria were met together: abnormal objective testing (Schirmer's test ≤10 mm or tear break-up time [TBUT] under 10 seconds), relevant diagnostic coding, and initiation of topical DED medication.
Visual disturbances and ocular pain, being subjective, required documentation on at least two separate visits to count.
Who was included in the study?
The final cohort was 48,892 eyes of 27,372 patients. Of those, 22,263 patients (81.3%) had LASIK and 5,109 (18.7%) had PRK.
- Eyes with prior refractive surgery, keratoconus or corneal ectasia, or fewer than 6 months of follow-up were excluded.
Demographics:
- Median age was 39 years for LASIK and 40 for PRK
- Women made up 57.4% of the LASIK cohort and 54.8% of the PRK cohort
- Bilateral surgery was more common with LASIK (81.9%) than PRK (64.3%)
- At baseline, a history of DED was present in 13.7% of LASIK eyes and 17.6% of PRK eyes
Findings?
DED was the most documented complication, presenting in 16.1% of LASIK eyes and 15.6% of PRK eyes over the full follow-up.
The PRK peak: During the 1 to 3 months window, DED was more common after PRK affecting 4.4% of patients compared with 3.7% after LASIK.
Visual disturbances: Although uncommon overall, visual disturbances were more frequent after PRK early in the postoperative period. During the first month, they occurred in 0.8% of PRK eyes compared with 0.3% of LASIK eyes. This difference diminished over time and was no longer significant at later follow-up visits.
Tell me more.
Between 3 and 12 months after surgery, dry eye was slightly more common after LASIK than PRK (6.1% vs 5.4%). However, the difference was not statistically significant, so it should be interpreted with caution.
Ocular pain: Was rare across the board, at 0.1% in both groups.
- The only notable difference occurred between 3 and 12 months postoperatively, with LASIK showing a lower risk than PRK.
- Earlier windows showed no consistent difference, and the 1 to 3 months comparison couldn't be estimated because no PRK pain events were recorded
In the pooled model: Female sex, geographic region, a prior history of DED or visual disturbances, and earlier surgical years were each associated with higher complication risk.
- A more recent surgery year was protective (IRR 0.9).
Limitations?
This is electronic health record (EHR)-based registry data, with the usual caveats.
And considering the IRIS Registry leans toward private practices and academic centers, it underrepresented high-volume corporate refractive centers that may document and follow patients differently.
The bigger constraint: Coding-based outcomes only captured complications severe enough to generate a clinical code, which means subjective symptoms like visual disturbances and ocular pain were almost certainly undercounted.
- The analysis also caught only the first documented occurrence of each complication, so it can't speak to whether symptoms persisted or became chronic.
Expert opinion?
No outside expert commentary was included in the study. The authors made their own case.
Their position: The timing differences should shape how clinicians counsel and follow patients. After PRK, front-load ocular surface support with aggressive lubrication and anti-inflammatory therapy to blunt the early dry-eye and visual-disturbance peak. After LASIK, extend DED surveillance past the early window, since symptoms there can linger.
On the pain numbers: The authors argued the very low coded rates likely reflect modern prophylaxis, including bandage contact lenses and multimodal analgesia, alongside a documentation issue..
PRK pain is expected and transient, so clinicians may not code it, while unanticipated LASIK pain is more likely to get logged.
Take home.
Procedure and timing both shape complication risk. PRK patients are more likely to hit early dry eye (1 to 3 months) and early visual disturbances (first month), so it's worth optimizing the ocular surface up front.
As such: LASIK patients warrant longer dry-eye follow-up, since their risk doesn't fully settle in the early window. And either way, registry-coded rates understate what patients actually report, so a careful symptom history still tells you more than the chart codes do.