To the Editor:

In the November 2006 issue, Dr. Ronald Gallemore undertook the comprehensive task of reviewing the current state of non-steroidal anti-inflammatory drugs in the treatment of ocular diseases ("NSAIDs in Treatment of Retinal Disorders", p. 81). Although he correctly stated that NSAIDs are approved for "the treatment of a variety of anterior segment conditions including postoperative inflammation following cataract surgery, prevention of miosis during cataract surgery, and postop pain following refractive and cataract surgery," he also appropriately addressed the reality of the expanding role of these medications to include numerous posterior segment diseases.

As Dr. Gallemore suggests, it is becoming more and more commonplace for ophthalmologists to be using NSAIDs—either alone or in combination with other agents or modalities—to treat conditions such as Irvine Gass syndrome, diabetic macular edema, cystoid macular edema from branch and central retinal vein occlusions, secondary macular edema associated with epiretinal membranes and even exudative age-related macular degeneration. The expanding utilization of this class of drugs, both for on and off-label applications, is at least in part a reflection of a growing body of evidence supporting this trend.

When discussing NSAID efficacy, Dr. Gallemore states that there have been "no large, randomized, controlled clinical trials comparing the efficacy of topical NSAIDs for the treatment of ocular disorders" and that "prospective, randomized controlled studies are required to assess the relative efficacy of NSAIDs for specific conditions." Although more such trials are necessary, to be sure, there were two studies presented at the 2006 American Academy of Ophthalmology (subsequent to the publication of Dr. Gallemore's article) that sought to evaluate the efficacy of NSAIDs in controlling intraocular inflammation and preventing pseudophakic cystoid macular edema.

The first study, "Aqueous concentration and PGE2 inhibition of Ketorolac 0.4% vs. Nepafenac 0.1% in Phacoemulsification Patients," by Frank Bucci, MD, was a masked trial involving 132 patients that were randomized to receive a preoperative regimen of ketorolac 0.4% or nepafenac 0.1% prior to undergoing phacoemulsification.1 At the time of cataract surgery, an aqueous specimen was collected. Data revealed that ketorolac appeared to have greater aqueous penetration that nepafenac and amfenac (its active metabolite) and that ketorolac-treated eyes were significantly more likely to have undetectable PGE2 levels (<100pg/mL), which is a significant source of intraocular inflammation.

The second study entitled "A Masked Comparison of Acular LS Plus Steroid vs. Steroid Alone for the Prevention of Macular Leakage in Cataract Patients" by John Wittpenn, MD, and colleagues was a randomized, investigator-masked, multicenter clinical trial with n=171/500.2 The group of patients that received steroid plus Acular LS was significantly less likely to develop macular thickening on OCT. The "steroid alone" group also had a greater incidence of clinical CME.

These are just two examples of well-designed clinical trials comparing various NSAIDs and assessing their relative effectiveness in suppressing inflammation and controlling the eventual macular edema that can result from this inflammatory stimulus. As retinal specialists and ophthalmologists in general continue to broaden the therapeutic applications of this potent class of drugs, it is imperative that these practices continue to be based on sound scientific research and evidence-based medicine. Nonsteroidal anti-inflammatory drugs have played a vital role in the history of modern ophthalmology as a result of a strict adherence to these principles. It is indeed this generation's responsibility to continue these practices.

Miguel A. Busquets, MD, FACS



1. Bucci F. Aqueous concentration and PGE2 inhibition of ketorolac 0.4% vs. nepafenac 0.1% in phacoemulsification patients. AAO Scientific Poster #382 2006.

2. Wittpenn J, et al. A Masked Comparison of Acular LS Plus Steroid vs. Steroid Alone for the Prevention of Macular Leakage in Cataract Patients. AAO Scientific Poster #5, 2006.


To the Editor:

I appreciate Dr. Busquets' interest in my review article. He concurs with the expanding roles of NSAIDS for managing posterior segment diseases and the need for additional studies evaluating such applications.

In his letter, he includes preliminary data reported after the publication of my article on the comparative efficacy of topical NSAIDs, including Dr. Bucci's study that reported that ketorolac had a greater aqueous concentration than nepafenac or amfenac. The study did not, however, demonstrate a statistical difference between the combination of nepafenac and its hydrolysis product, amfenac; the combination reflects the true penetration of the drug into the eye. In addition, patients received a dosing regimen that is not FDA-approved, with the administration of four drops of an NSAID 90 minutes prior to surgery. The final mean PGE2 measurement shows that the ketorolac group had a lower PGE2 level, but the relative reduction of PGE2 from baseline was not reported, so the percentage of inhibition cannot be compared.

Dr. Busquets also reports the preliminary data of Dr. Wittpen et al, analyzing postop CME with and without the addition of Acular LS to steroids. This study supports the adjunctive role for NSAIDs in the treatment of CME. Another preliminary study supporting NSAIDs in CME was presented by David Rho, MD, at the 2006 AAO and demonstrated comparable efficacy of ketorolac q.i.d to bromfenac b.i.d.1 The lower dosing frequency of bromfenac over diclofenac may impact drug selection in such cases.

The roles of potency and penetration are also highlighted in Dr. Busquets' letter. An NSAID's potency appears to correlate most directly with its ability to inhibit the COX-2 enzyme, the key modulator of ocular pain and inflammation. Until recently, the IC50s for COX-2 have been measured using different protocols. A new study reports IC50s for the major NSAIDS using the same protocol for each with the relative potencies as follows: bromfenac (IC50=7.5nM) > amfenac (20.4 nM) > ketorolac (27.9 nM) > diclofenac (30.7 nM).3

We look forward to seeing how the relative COX inhibition values compare to the relative efficacy of NSAIDs in the treatment of posterior segment disorders. Larger randomized studies of NSAIDs will be needed to assess these differences.

Ron P. Gallemore, MD, PhD

Beverly Hills, Calif.


1. Rho DS. Treatment of acute pseudophakic cystoid macular edema: Bromfenac 0.09% versus diclofenac sodium 0.1% versus ketorolac tromethamine 0.5%. AAO Scientific Poster #314, 2006.

2. Waterbury DL, Silliman D, Jolas T. Comparison of cyclooxygenase inhibitory activity and ocular anti-inflammatory effects of ketorolac tromethamine and bromfenac sodium. Current Med Res Opinion 2006;22:1133-40.

3. Kida T, Ogawa T, McNamara TR Song CK, Gow JA Evaluations of the Human COX-2 Inhibition for Amfenac, Bromfenac, Diclofenac, and Ketorolac. ASCRS abstract, 2007.