Early complications after laser eye surgery — the clock, and the flight
Published evidence researched September 2026 · Prices researched July 2026 · Medically reviewed
Two different things can happen in the space under a LASIK flap in the first fortnight, they look nearly identical at the slit lamp, and they are treated in opposite directions. Diffuse lamellar keratitis is inflammation and wants more steroid. Pressure-induced stromal keratopathy is caused by the steroid and wants it stopped. Worse, the instrument that should separate them — central applanation tonometry — reads falsely low over a pocket of interface fluid, so the reassuring number is manufactured by the problem itself. The part that is specific to flying in for surgery, and that this page exists to state: DLK is staged by how far it has migrated across the cornea, so its stages are a calendar — stage 1 on days 1–2, the threshold stage after further inward migration over the days that follow. A fly-in laser itinerary runs about five to seven days. The easily-treated stage happens while you are still in China; the stage that scars can arrive as you board.
An interface is a space that did not exist before your operation
Every complication on this page except one depends on a single fact of anatomy that is easy to miss: lamellar refractive surgery creates a space inside the cornea that was not there before. A femtosecond laser separates a flap, the treatment is applied to the bed beneath it, and the flap is laid back down. It seals by apposition rather than by stitching, and the plane where it meets the bed is the interface. In a healthy unoperated cornea there is nothing corresponding to it — no potential space, nothing that can fill, nothing that can collect.
That space is why the two conditions this page is mostly about exist at all, and it is why they are hard to tell apart. Both of them announce themselves as a haziness in the same plane. One is inflammatory cells that have arrived there; the other is fluid that has been driven there. Looking at a grainy interface through a slit lamp does not, by itself, tell you which. And the treatment for one makes the other worse.
It also explains the sharpest thing in this page's procedure table below: the risks are not degrees of a single hazard that you take more or less of depending on the operation. They are different hazards belonging to different operations, because the operations leave the cornea in structurally different states.
DLK: the staging system is also a calendar
Diffuse lamellar keratitis is a sterile inflammatory reaction in the interface — sterile meaning it is not an infection, which is why steroid rather than antibiotic is the mainstay of treatment. The syndrome was described under that name by Smith and Maloney in Ophthalmology in 1998, and the nickname sands of the Sahara comes from the grainy white appearance of the cells in the plane. Incidence in the LASIK literature is commonly quoted around 2–4% of eyes, and that range is wide for an honest reason: series differ in how closely they examined asymptomatic eyes and in what they counted.
The conventional description divides it into four stages, and the thing worth internalising is that the stages are defined by distance travelled across the cornea, and travelling takes days. The staging system therefore doubles as a timetable, which is what makes it possible to lay it against a flight.
| Stage | Typically appears | What is happening in the interface | What it means for vision |
|---|---|---|---|
| 1 | Days 1–2 | Inflammatory cells in the periphery only, sparing the centre | Usually none noticed. Commonly found at a scheduled check rather than reported |
| 2 | Days 3–4 | Cells migrating inward toward the visual axis | Acuity begins to be affected; the change may be what prompts the call |
| 3 | After further inward migration | Central involvement; clinicians call this the threshold stage | Permanent scarring and measurable loss of vision become likely, not merely possible |
| 4 | Later still, uncommon | Stromal melting and further scarring | Can leave a lasting shift in the prescription |
Staging as described in the standard refractive-surgery literature, researched September 2026. Stage boundaries are clinical judgements made at a slit lamp, not dates on a chart — the days given are the typical pattern, and an individual eye can be faster, slower, or can stop on its own at stage 1. Treatment response is why the later stages are uncommon: all stages typically respond to corticosteroid, and at the higher stages lifting the flap and irrigating beneath it is part of the repertoire.
★ Set that clock against the trip, because they are the same clock
Here is the observation this page was built to make, and as far as we can find nobody states it anywhere: the DLK staging clock and the medical-tourism itinerary run on the same days, and they are pointed in opposite directions.
The laser itineraries this site publishes run about five to seven days, because departure waits on the post-operative checks rather than on how comfortable you feel. Overlay the two and the arithmetic is plain. Stage 1 — the stage that is trivial to treat, and the stage that most often produces no symptoms at all — falls squarely inside your stay, while you are still in the same city as the surgeon who operated and still attending checks at the clinic that has your file. Stage 3, the threshold stage, is reached by migration over the following days — which is to say around the time you are packing, boarding, or in the air.
Two conclusions follow, and neither is the alarming one.
The first is that the value of the in-country checks is much higher than it looks, because their real function is to catch something you cannot feel during the only window in which catching it is easy. A patient who skips the last check because their vision seems fine has misunderstood what the appointment is for. Early DLK is frequently asymptomatic; feeling fine is not evidence.
The second is that the flight date is a clinical variable and should be treated as one. Not moved by default — the itinerary is set by clinical gates and lengthening it by two days would not cover the whole tail anyway — but treated as something that can move if the surgeon at the last check says the interface does not look entirely quiet. The cheapest moment to buy a flexible ticket is before anything has happened. Against procedure costs of $1,200–1,800 for LASIK and $1,000–1,500 for TransPRK (researched July 2026, consistent with this site's price table), a change fee is a rounding error, and it is the only form of insurance on this page that you can buy in advance.
★ PISK: the lookalike whose treatment runs the other way
If you read one section here, read this one, because it is the only place on the page where a well-intentioned treatment decision can make things materially worse.
The steroid drops given after surgery to suppress inflammation raise intraocular pressure in some people. Raise it enough and fluid is driven into the interface — the same space DLK occupies. The result is called pressure-induced stromal keratopathy, PISK, first reported by Lyle and Jin in 1999 and also written as pressure-induced interlamellar stromal keratitis. At the slit lamp, early PISK and DLK can look nearly identical: a diffuse haziness in the interface may be the only sign either produces.
The instrument that should settle it reads the wrong way
Eye pressure is normally measured by applanation: a small central area of cornea is flattened and the force needed is read off. A pocket of fluid sitting in the interface behaves like a cushion. It absorbs the applanating force rather than transmitting it, and the instrument reports a pressure well below the true one. Case reports describe eyes running dangerously high pressures while central readings came back comfortably normal.
Sit with the shape of that for a moment, because it is unusual and it is the reason this page insists on paperwork. The reassuring number is manufactured by the condition it is failing to find. This is not a test that is merely insensitive; it is a test whose error runs in the dangerous direction, and it runs that way because the disease is present. The clinical answers are to measure away from the centre, over peripheral cornea outside the fluid pocket, and to image the plane directly with anterior-segment OCT, which shows a fluid space that no pressure reading will reveal.
None of which a patient can do anything about at the moment of examination — but one thing makes the peripheral check far more likely to happen, and it is entirely within your control: the examining doctor knowing that you are on a post-laser steroid, at what strength, and for how long. That single fact is what turns a normal-looking pressure into a question. It is the reason the taper schedule leads the document list below.
The procedure split: different operations, different early risks
It is tempting to think of these as one pool of risk that the three laser procedures draw from in different amounts. That is not what the comparative data shows. The early complications are substantially procedure-specific, because they depend on what the operation leaves behind. A 2026 comparative analysis of the three techniques is unusually clean on this point: transient corneal haze was reported exclusively in the TransPRK group, at 3.2% at three months, while DLK and epithelial ingrowth were reported exclusively in the femtosecond-LASIK group.
| Procedure | What it leaves in the cornea | Characteristic early problem | Timescale of the worry |
|---|---|---|---|
| LASIK | A hinged flap over a full-diameter interface | DLK, PISK, flap striae, flap displacement, epithelial ingrowth | Days. The first fortnight carries nearly all of it |
| SMILE | No flap; a pocket reached through a small incision | Interface events are described but are rarer; the small opening is less exposed | Days, with a lower overall interface burden |
| TransPRK | No flap and no interface — a treated surface healing under a bandage lens | Haze, and a slower, genuinely sore epithelial healing course | Weeks to months rather than days |
| EVO ICL | A lens inside the eye — not a corneal operation at all | A different category entirely: infection inside the eye is the defining risk | Covered separately — see the ICL page linked below |
Comparative figures as reported in a 2026 three-way analysis of SMILE, femtosecond LASIK and TransPRK, researched September 2026. Single-study figures describe one series and one set of definitions; read them as the shape of the difference between the procedures rather than as your own probability. Which procedure is appropriate for your eyes is determined by the operating surgeon after the diagnostic workup, and the risk profile above is one input to that decision among several.
The mechanical flap problems, briefly, because they are LASIK's alone
Three things can happen to a flap that cannot happen to a surface treatment or to a pocket. Striae are wrinkles in the flap; fine ones are common and inconsequential, while coarse ones crossing the visual axis distort vision and are refloated and smoothed if they matter. Displacement — the flap physically shifting — is overwhelmingly a first-days event and is overwhelmingly caused by rubbing or an impact, which is the entire reason for the eye shield you are told to wear at night and the instruction not to touch the eye. It is the one complication on this page whose main risk factor is under your direct control. Epithelial ingrowth is surface cells creeping under the flap edge; it appears over weeks rather than days, is more likely where the epithelium was disturbed during surgery, and is managed by lifting and cleaning the edge when it advances.
The traveller's angle on all three is the same and is worth stating once: they are mechanical, so they announce themselves quickly and visibly, and unlike DLK they rarely hide. If your vision was good on day two and is distinctly worse on day three with a gritty, something-is- wrong sensation, that is a flap conversation, and it is a today conversation.
What you should have in writing before you fly
The premise is unglamorous: you are the only continuous record of your own care. The surgeon who operated and the doctor who sees you at home will probably never speak to each other, and the second one cannot infer from looking at your eye what the first one did. Ask for this in English before you leave, and do not settle for a discharge note that names the procedure and stops.
- The steroid, its strength, and the taper written as dates. First on the list for the reason given above: it is the fact that makes the DLK-or-PISK question askable. “Week two” is not good enough — a doctor meeting you on a Thursday needs calendar dates.
- Every other drop, with strength and frequency, including the lubricants.
- The exact procedure and the platform it was performed on, by name and model.
- Whether there was any epithelial defect during the operation. This is the item people never think to ask for, and it changes your risk substantially rather than marginally: a published series reports a markedly elevated risk of DLK — an order of magnitude and more — in eyes with an epithelial defect after LASIK compared with controls. If your operative note records one, that is a reason to discuss the flight date rather than assume it.
- Pre-operative measurements, including corneal thickness and the instrument that produced it, since thickness figures are not interchangeable between devices.
- The post-operative pressure readings taken before you left — a baseline is what lets a doctor at home judge whether today's number has moved.
- The date of surgery and which eye, if they were not done together.
- A working after-hours route back to the operating team — a number or an address that is answered outside office hours, tested before you need it.
Arranging the second half of the window before you need it
An optometrist is a sensible first contact and is usually the appointment you can get quickest, but what this page is about needs a slit-lamp examination of the interface by someone who can act on the finding — an ophthalmologist with corneal or refractive experience. The limitation is information and equipment, not competence: a clinician meeting you cold does not know your procedure, your platform, your operative note or your taper, and may not have the OCT that settles the question.
So do the boring thing in advance, while nothing is wrong. Identify a corneal or refractive ophthalmologist you could reach within a day or two. Find out whether your system lets you self-refer, or whether you need to go through a GP first and how long that takes. Find out what happens to an eye problem in your area outside working hours — which service, where, and whether it has an ophthalmologist on site or on call. Then put those details in the same place as the paperwork above. Done in advance this takes twenty minutes and converts an emergency into an appointment; done during a bad week it is the hardest thing on the list.
Three patterns that mean today rather than at your next appointment
Any one symptom in isolation is usually nothing, which is why these are written as patterns.
- Vision with a direction of travel. Ordinary recovery fluctuates — clearer in the morning, hazier at night, better tomorrow. What matters is two or three consecutive days each worse than the last.
- Pain that is increasing rather than settling. Grittiness and light sensitivity that ease over days are expected. Worse on day four than on day two is not — and after a surface procedure this holds even inside the sore period you were warned about.
- One eye diverging from the other after they had been tracking together. Both eyes had the same operation on the same day, which makes them the best controlled comparison you have. When one falls behind, that asymmetry is information.
None of those tells you what is wrong, and neither does this page. What they tell you is that the timing question has been settled: you need looking at now rather than on schedule.
What this page deliberately does not do
It publishes no complication rate for any named hospital. Single-centre rates rest on small denominators, are self-reported by the party with an interest in them, and cannot be audited by a patient — a centre quoting a very low or zero rate is usually quoting a small number of operations, which is the opposite of reassuring. What is worth asking about is process, because process is observable: the post-operative check schedule and who performs it, whether anterior-segment OCT is available on site, how an after-hours presentation is handled, and what happens if a check finds something the day before your flight.
It also does not tell you your own risk, and nor can anything else you read. Published incidence describes populations assembled under particular definitions over particular periods. Yours depends on your cornea, your tear film, your healing and what happens during your operation. Candidacy for any procedure is determined by the operating surgeon after the diagnostic workup — which is a statement about who holds the information, not a disclaimer.
Questions people actually ask
What is diffuse lamellar keratitis, and how soon after laser eye surgery does it appear?
Diffuse lamellar keratitis — DLK, and informally “sands of the Sahara” for the grainy appearance it produces — is a sterile inflammatory reaction in the space between the corneal flap and the bed beneath it. Sterile is the important word: it is an inflammatory response, not an infection, which is why it is treated with steroid rather than with antibiotic, and why it is a different problem from the infections described elsewhere on this site. It is a complication of lamellar surgery specifically, because it needs an interface to happen in, and an interface is a space that does not exist in an eye that has not been operated on. The classic presentation begins one to two days after surgery. Published descriptions of the natural history go back to Smith and Maloney's 1998 account in Ophthalmology, which named the syndrome; incidence in the LASIK literature is usually quoted in the region of 2–4% of eyes, a figure that depends heavily on how hard a given series looked and on what it counted as a case. Early DLK frequently produces no symptoms you would notice, which is the single most important thing to understand about it: it is often found at a scheduled slit-lamp check rather than reported by the patient. That is why the post-operative appointment schedule matters more than how you feel.
Why does the staging of DLK matter so much to someone who flew in for surgery?
Because the stages are defined by how far the inflammatory cells have migrated across the cornea, and migration takes days, so the staging system is also a calendar. In the conventional four-stage description: stage 1 appears around days 1–2 and sits in the periphery, away from the centre of your vision; stage 2 typically arrives around days 3–4 as cells move inward and begin to affect acuity; stage 3 is the one clinicians call the threshold, because at that point permanent scarring and a measurable loss of vision become likely rather than merely possible; stage 4 involves stromal melting and can leave a lasting shift in the prescription. Now lay the fly-in itinerary over it. The laser trips this site publishes run about five to seven days, because departure waits on the checks rather than on how you feel. So the stage that is genuinely easy to treat is the one that happens while you are still in the same city as your surgeon, and the stage at which the damage becomes permanent is the one that can arrive around the time you are boarding. That overlap is the whole reason this page exists. It is not an argument against travelling for surgery, and the itinerary is not wrong; it is an argument for knowing, before you fly, exactly who looks at your eye on which day.
What is PISK, and why is it the most dangerous thing on this page?
Pressure-induced stromal keratopathy — PISK, first described by Lyle and Jin in 1999, and also written as pressure-induced interlamellar stromal keratitis — is what happens when the steroid drops you were given to prevent inflammation raise the pressure inside your eye enough that fluid is driven into the interface. It is dangerous for one specific reason: at the slit lamp it can look almost exactly like DLK, and the two are treated in opposite directions. DLK is inflammation, so the response is more steroid. PISK is caused by the steroid, so the response is to reduce or stop it and bring the pressure down. Treat PISK as DLK and you accelerate the thing you are trying to stop, while the underlying pressure goes on damaging the optic nerve. The literature is explicit that early PISK is misdiagnosed as DLK and that this delays correct treatment. This is not a criticism of anyone's competence — it is a genuinely difficult distinction that experienced corneal surgeons write papers about. It is, however, a decisive argument for your steroid schedule being written down and carried, because it is the single piece of information that makes the distinction askable.
Why can't the eye pressure reading be trusted in that situation?
This is the part that most surprises people, and it is worth understanding rather than just accepting. The standard way of measuring eye pressure is applanation tonometry, which flattens a small area at the centre of the cornea and reads the force required. When there is a pocket of fluid sitting in the interface, that pocket behaves like a cushion: it absorbs the applanating force instead of transmitting it, and the instrument reports a pressure substantially lower than the true one. Case reports describe eyes whose real pressure was dangerously high while central readings came back in the normal range. So the reassuring number is not merely unhelpful, it is produced by the condition it is failing to detect — a false negative generated by the disease. The practical answers clinicians use are to measure away from the centre, over the peripheral cornea outside the fluid pocket, and to image the interface directly with anterior-segment OCT, which shows a fluid space that no pressure reading will reveal. As a patient the useful move is not to argue about tonometry, it is to make sure the person examining you knows you are on a post-laser steroid and how long you have been on it. That one fact is what prompts the peripheral check.
I had TransPRK, not LASIK. Does any of this apply to me?
Much less of it, and the reason is structural rather than reassuring. TransPRK is a surface procedure: the epithelium is removed and the treatment is applied to the surface of the cornea, so no flap is cut and no interface is created. Almost everything on this page needs an interface to happen in, so DLK in its classic form, epithelial ingrowth and flap displacement are not the risks that belong to your procedure. What belongs to yours instead is the surface healing course, and its characteristic problem is haze — a loss of corneal clarity that develops over weeks rather than days. A 2026 comparative analysis of the three techniques found transient haze reported exclusively in the TransPRK group, at 3.2% at three months, while DLK and epithelial ingrowth were reported exclusively in the femtosecond-LASIK group. That split is the useful thing to take away: these are not degrees of the same risk, they are different risks attached to different operations. Your early course is also more uncomfortable and slower, with a bandage contact lens and several days of genuine soreness, and that is expected rather than a complication. SMILE sits between the two — no flap, but a pocket, so interface events are described in the literature and are rarer.
What should I actually be watching for once I have gone home?
Three patterns, and the reason to learn them as patterns rather than as symptoms is that any single symptom on its own is usually nothing. The first is vision that is going backwards. Normal recovery fluctuates — clear in the morning, hazier by evening, better again the next day — and that is not what this means. This means a clear direction of travel over two or three consecutive days, in which each day is worse than the one before. The second is pain that is increasing. Discomfort, grittiness and light sensitivity that gradually settle are ordinary; pain that is worse on day four than on day two is not, and after a surface procedure this warrants contact even within the expected sore period. The third is one eye diverging from the other after they had been tracking together, because your two eyes had the same surgery on the same day and are the best controlled experiment available to you. Any of those three is a reason to be examined, not a reason to wait for your scheduled appointment. None of them tells you what is wrong, and this page cannot either — they tell you the timing question has been answered.
Who should I see at home, and is my optometrist enough?
An optometrist is a reasonable first contact and is often the fastest appointment available, but the examination you need is a slit-lamp look at the interface by someone who can act on what they find, and an ophthalmologist with corneal or refractive experience is the right destination. The specific limitation is not skill, it is information and access. A clinician meeting you for the first time does not know which procedure you had, what was used, whether anything was noted during the operation, what you are dropping or on what taper — and none of that is inferable from looking at your eye. They may also not have anterior-segment OCT, which is what settles the question this page is really about. The genuinely useful step, and it costs nothing, is to identify that person before you fly rather than during a bad week: find a corneal or refractive ophthalmologist near you, check what a self-referred appointment requires, and find out what your local system does with an eye problem outside working hours. Arranging it in advance turns an emergency into an appointment.
What do I need to take home with me in writing?
The principle is that you are the only continuous record of your own care, because the surgeon who operated and the doctor who sees you at home may never speak. Ask for it in English before you leave, and do not accept a discharge summary that only names the procedure. What is worth having: the exact procedure and the platform it was performed on; the date and which eye was treated first if they were not simultaneous; the pre-operative measurements, including corneal thickness and the instrument that produced it; anything noted during the operation, and in particular whether there was any epithelial defect, because a defect raises the DLK risk substantially rather than marginally; every drop you are on, with its strength and the taper schedule written as dates rather than as week numbers; the post-operative pressure readings taken before you left, which give a doctor at home a baseline to compare against; and a working after-hours route back to the operating team. The taper is the one people omit and the one that matters most here, for the reason set out above: without it, the question that separates the two conditions on this page cannot be asked.
Does this mean having laser eye surgery abroad is a bad idea?
No, and a page about complications is a bad instrument for answering that question, because it is written entirely in the vocabulary of what can go wrong. What can be said honestly is narrower and more useful. These complications are uncommon, most resolve completely when they are found and treated in time, and none of them is specific to China or to travelling — the same list applies to the same operations performed anywhere. What travelling changes is not the risk, it is the geometry of follow-up: the window in which a problem is cheap to fix overlaps the window in which you are moving between countries. That is a planning problem, and planning problems have planning answers — knowing the staging clock, carrying the paperwork, identifying a doctor at home in advance, and treating the flight date as a clinical variable rather than a fixed one. Set against the procedure costs on this site's price table — LASIK at $1,200–1,800 and TransPRK at $1,000–1,500, researched July 2026 — the cost of a changed flight is small, and it is the cheapest piece of insurance available to anyone who needs to use it.
Can you tell me my own risk of these complications?
No, and neither can any page that offers to. Published incidence figures describe populations, not people, and they come from series that differ in how they defined a case, how closely they looked and over what period they collected — which is why the ranges in the literature are as wide as they are. Your own risk depends on your cornea, your tear film, your healing, which procedure is appropriate for you and what happens during the operation, and the only person in a position to weigh those is the surgeon who has examined you and read your scans. Candidacy for any procedure here is determined by the operating surgeon after the diagnostic workup, and that is not a legal formula — it is a statement about who holds the information. What this page is for is different and narrower: it is about what the calendar does to a complication once one has occurred, which is the part that is genuinely specific to travelling for surgery and the part that is almost never written down.