Corneal Infiltrate and Keratitis

I like to keep my notes according to what I will see in clinic

What are the causes of a corneal infiltrate?

AAO PPP

  • Microbial keratitis

    • Bacterial, most common includes staphylococcus (typically coagulase negative) and pseudomonas (typically aerginosa)

    • Fungal

    • Acathamoebal

    • Nontuberculous mycobacterium

  • Non infectious (aka sterile) stromal infiltration (produced by antigens from local or systemic bacterial infection)

    • Antistaphylococcus associated marginal keratitis

    • Peripheral ulcerative keratitis secondary to autoimmune disease

What features are characteristic of what differential?

Microbial keratitis in general

AAO PPP

Risk factors of microbial keratitis include:

  • Recent trauma (e.g. foreign body, corneal laceration, abrasion, surgery etc.)

  • Existing or recent herpetic (HSV/VZV) disease (e.g. bacterial superimposition)

  • Contact lens wear (the most common risk factor), particularly with poor hygiene (see “counselling on contact lens hygiene”)

  • Ocular surface disruptance: dry eye disease (ant/post blepharitis, lagophthalmos, ectropion, punctual ectropion, etc.); RCES, etc.

Features

  • Corneal infiltrate +/- epitheliai defect

  • +/- anterior chamber reaction

  • +/- else

Bacterial keratitis

AAO PPP

Specific risk factors:

  • Contact lens wear -> ? pseudomonas (check this, this is from memory)

  • LASIK -> nontuberculous mycobacterium

  • Trauma -> polymicrobial keratitis (~40% of scrapes may yield two or more organisms)

Features

  • Infiltrate of the corneal stroma, > 1 mm in size, with white cell infiltration and corneal oedema in the surrounding stroma, hazy margins

  • +/- epithelial defect +/- antrior chamber reaction

  • +/- hypopyon

    • Hypopyon in bacterial keratitis is typically sterile

Fungal keratitis

AAO PPP

fungal includes yeast and mold

Risk factors

  • Plant material (double check)

Features

  • infiltrate appears dry instead of suppratative

  • feathered edges

  • satellite lesions

  • posterior plaque

  • A ring lesion is seen in both fungal and acanthamoebal keratitis

A study showed that corneal specialists can visually differentiate between bacterial and fungal keratitis in around 70% of the time (ref below).

Acanthamoebal keratitis

AAO PPP

protozoal parasite

Features:

  • Inflammation along corneal nerves (called radial keratoneuritis)

  • Pain out of proportion of findings

  • A ring lesion is seen in both fungal and acanthamoebal keratitis

Marginal keratitis

What is this, and what is its typical appearance?

Peripheral ulcerative keratitis

What is this, and what is its typical appearance?

When to scrape vs not to scrape?

AAO PPP

here is the checklist:

  • Hx of ocular surgery involving the cornea

  • Hx of organic matter (ie gardening etc.)

  • Presence of >= 1+ cells in the AC

  • The infiltrate is >/= 2 mm AND within 3 mm of corneal centre

  • Multiple (2 or more) lesions

  • Significant stromal involvement or melting

  • Atypical features (ie suggestive of fungal, amoebic, or mycobacterial)

  • Poor response to empirical treatment

How to treat?

What drug to prescribe?

Topical antibiotics

Jeremy Williamson

  • Topical Cefazolin is for strep cover

AAO PPP - ocular ointments lack solubility, and so are not able to penetrate into the cornea for optimum therapeutic effect. They may be used for adjunctive therapy.

Single drug therapy using fluoroquinolone has been shown to be as effective as combination therapy antibiotics

fortified topicals should be reserved for large infiltrates, visually significant infiltrates, or if hypopyon is present

Single therapy agents include ciprofloxacin 0.3%, ofloxacin 0.3%, and levofloxacin 1.5% in america (FDA)

Consider risk factors for fluoroquinolone resistance (e.g. recent fluoroquinolone use, or realistically, if they use cipro or so long term)

MRSA has high resistance to fluoroquinolones

A cochrane review found no evidence in the rates of corneal perforation between the different antibiotics

you give fortified if sever or unresponsive

pharmacy should prep them

Topical cyloplegia

If there is anterior chamber reaction with bacterial keratitis, cycloplegia may be useful to decrease synechiae formation and decrease pain

PPP

Topical steroids

AAO PPP

a number of studies show there is no difference in clinical outcome with the addition of corticosteroids

there is a theoretical advantage from suppression AC information, which may reduce corneal scarring

There are theoretical risks of recurrence of inection, inhibition of collagen synthesis (predispoing to corneal melt) and increasing IOP

The steroids for corneal ulcer trial foud no benefit of concurrent steroid use with broad spectrum topical antibiotc. But also, this study did not find an increase in the adverse effects in people with bacterial keraetitis (so it sounds here and there).

subgroup analysis revealed a potential benefit in using corticosteroids in two populations:

  • pseudomonas keratitis

  • very severe ulcers of bacterial keratitis, ie covering the central 4mm pupil, or VA with CF or worse)

corticosteroids worsened nocardia keratitis

In non-nocardia keratitis, anotehr study foud that addition steroids within 2-3 days of antibiotic therapy (compared to 4 or more days later) resulted in 1 line better.

no clear cut answer

A conservative approach is to avoid prescribing steroids until the organism has been identified, the ED is healing and/or the ulcer is consolidating.

If the ulcer is associated with nocardia or fungus, the outceomes are likely to be poor. There was initial data that showed initial CS use in fungal keratitis had a higher risk for requiring penetrating keratoplasty. Although a more recent trial didn ot find this, still high risk, who knows.

So it seems that:

  • you must exclude fungal or nordic keratitis before considering steroid therapy

  • the epithelial defect and the stromal infiltration must both be healing

  • the benefit of adding corticosteroid is more pronounced when commenced before 4 days, but not at the expense of the above obviously

  • remember, this is in the presence of AC reaction

if you’re going to start steroid therapy, you shold see them within 2 days of commencing.

make sure to monitor IOP

patients already on corticosteroid for other reasons need to sspend therapy until infection controlled and organism identified.

Frequency and weanne?

For central keratitis (defined earlier as within 3 mm of the centre) or for severe keratitis (defined earlier as > 2 mm with extensive suppuration) a loading dose e.g. every 5 to 15 minutes followed by frequent applications every 1 hour is recommended.

  • AAO PPP

Manasi said this is Dr McLintock’s regime q1h for 48h then q1h awake for 5 days Q2h 1 week 6x/d 2 weeks

Prof Jeremy - q1h 48 hours is pretty standard - The rest is guided by clinical response

When to see again?

AAO PPP

severe cases = ie deep stromal involvement, or infiltrates larger than 2mm with extensive suppuration shold be followed daily at least until stable or clinical improvement confirmed

pseudomonas is known to produce increased inflammation int he FIRST 24-48 hours despite appropriate therapy

Here are markers of a positivce response:

  • decreased pain

  • decreased discharge

  • decreased injection

  • sharper demarcation of the perimetre

  • decreased densitty of the infiltrate without stromal loss

  • decreased stromal oedema

  • decrease in AC cells or hypopyon

  • ED healing

  • cessation of corneal thinning if present

you need to taper antibiotics. topical antibiotics cause toxicity

How do you deal with the poor responder? When to re-scrape?

AAO PPP

if there was a negative culture and lack of clinical response, then you can rescrape

do not confuse toxocity of medications or from corticosteroid withdrawal from worsening (this is interesting in people who are already taking steroids before and were susended)

consider discontiuing abx for 12-24 hrs pre reculture,it may increase yoeld

drops containing preservatives (like anaesthetic or cycloplegic agents) shold be avoided

consider other organisms like acanthamoeba, fusariu, atypical mycobacteria

Discharging

When is it safe to discharge?

…

Counselling on good contact lens hygiene

source: bacterial keratitis PPP

  • Always buy/use official lenses, never online from some dodgy store

  • Never share lenses with anyone

  • Do not wear them at night

  • Do not wear them for longer than the intended lifespan

  • Never clean them or store them in tape water

  • Wash your hands before putting them on or off

  • Make sure their container is clean

  • Never go swimming, shower with, or even to a hot tub/sauna/spa etc. with them on

old

Good resources, poached from Yanoff’s Ophthalmology

Yanoff Ophthalmology

O’Brien TP, Maguire MG, Fink NE, et al. Efficacy of ofloxacin vs cefazolin and tobramycin in the therapy for bacterial keratitis. Report from the Bacterial Keratitis Study Research Group. Arch Ophthalmol. 1995;113:1257–1265.

Srinivasan M, Mascarenhas J, Rajaraman R, et al. Corticosteroids for bacterial keratitis: the Steroids for Corneal Ulcers Trial (SCUT). Arch Ophthalmol. 2012;130:143–150.

Dalmon C, Porco TC, Lietman TM, et al. The clinical differentiation of bacterial and fungal keratitis: A photographic survey. Invest Ophthalmol Vis Sci. 2012;53:1787-1791.

  • in this study, 15 corneal specialists were only able to distinguish between bacterial and fungal keratitis in aroud 70% of the time. this shows the importance of getting a scrape. 70% seems pretty good, but not perfect.

Labetoulle M, Frau E, Offret H, et al. Non-preserved 1% lidocaine solution has less antibacterial properties than currently available anaesthetic eye-drops. Curr Eye Res. 2002;25:91-97. - apparently this study showed that topical anaesthetic decreases bacterial yield

cite the AAO PPP for the following table:

extra notes from foreign body keratitis

Always measure the ED and the stromal reaction Always check the fornices well with white light and again with fluorescein Always sweep the fornices Always notify the bosses Always do a posterior segment exam, looking for what??? - penetrating eye injury, but seidel neg, no ac reaction

Always ask if they are contact lens wearer, because that increases the risk of pseudomonas, which ocuflox doesn’t cover (or does it? ocuflox is ofloxacin, which does cover pseudomonas no?)

always document the depth of the infiltrate and the size

References

Bacterial keratitis PPP

See Tolestar and Ario Wilson pogmore article