Secondary Logo

Journal Logo


The role of perioperative nonsteroidal anti-inflammatory drugs use in cataract surgery

Brandsdorfer, Ari; Patel, Sejal H.; Chuck, Roy S.

Author Information
Current Opinion in Ophthalmology: January 2019 - Volume 30 - Issue 1 - p 44-49
doi: 10.1097/ICU.0000000000000541
  • Open Access



Nonsteroidal anti-inflammatory drugs (NSAIDs), a class of medications that inhibit cyclooxygenase (COX) enzymes from producing pro-inflammatory prostaglandins, have multiple indications in the perioperative period including pain control, reducing inflammation, improving intraocular mydriasis, and preventing postoperative cystoid macular edema (CME). Although the potential benefits of NSAIDs in the perioperative period have been known for years, optimal NSAID medication and dosing regimens have yet to be established. This article reviews the literature concerning perioperative NSAID use between 1 January 2016 and 30 June 2018.

Box 1
Box 1:
no caption available


Many cataract surgeons begin NSAID medications for their patients preoperatively to reduce intraoperative prostaglandin release and blunt postoperative inflammation. In 2017, Katsev et al.[1▪] obtained serial 100 μl samples of aqueous humor in 12 patients undergoing cataract surgery. Samples were obtained from patients who had received topical ketorolac three times over 24 h prior to surgery and were collected both after initial paracentesis, and prior to corneal wound hydration. Ketorolac concentrations were significantly reduced at the end of surgery compared with the beginning of surgery (P = 0.0022) with 66.7% of patients having undetectable levels at the end. Although the possible depot effect of preoperative ketorolac dosing in the vitreous cavity remains unclear, the authors noted that ketorolac's short half-life of 2.3 h and the removal of free drug by intraoperative irrigation likely renders the postoperative anti-inflammatory effects of preoperative ketorolac minimal. Therefore, they advocated for intraoperative ketorolac use.

Medic et al.[2] studied interleukin 12 (IL-12) concentration in the aqueous humor of 27 patients with diabetic retinopathy taking diclofenac 0.1% four times daily (Naclof, Exelsion, France) for 7 days prior to cataract surgery to 30 days following surgery compared with 28 patients who did not receive any perioperative NSAIDs. Samples taken after initial paracentesis found IL-12 concentrations were significantly reduced in the diclofenac group compared with control (33.4 ± 26.5 versus 57.7 ± 29.9 pg/ml; P = 0.007). Following surgery, patients in the diclofenac group had significantly lower central macular thickness (CMT) measurements by spectral OCT (Copernicus, OPTPOL Technology, Zawiercie, Poland). However, there was no difference in incidence of CME evident on OCT imaging between groups (37% treatment versus 43% control; P = 0.87).


Intraoperative miosis because of prostaglandin release from surgical insult can reduce a surgeon's visualization and lead to intraoperative complications. Omidira, a 1%/0.3% phenylephrine/ketorolac intraoperative injection (Omeros Corporation, Seattle, Washington, USA) added to irrigating fluid during cataract surgery was Food and Drug Administration (FDA)-approved in 2014 for prevention of intraoperative miosis and reduction of postoperative pain. Bucci et al.[3] demonstrated a significant decline in the use of pupil-expanding Malyugin Rings (MicroSurgical Technology, Redmond, Washington, USA) during uncomplicated phacoemulsification surgery with Omidira compared with balanced salt solution (BSS; 2.95 versus 7.87%; P < 0.001). Further sub-analysis of patients predisposed to intraocular floppy iris syndrome (IFIS) because of a history of alpha 1 blocker use demonstrated significantly fewer patients required Malyugin Ring use in the Omidria cohort compared with BSS (12.74 versus 24.5%, P = 0.05). Additionally, no patients in the Omidria cohort who underwent femtosecond laser capsulotomy, a process previously shown to cause prostaglandin release and intraoperative miosis, required Malyugin Ring use compared with 2.65% of control patients. Although the decreased need for pupillary expanders during femtosecond laser-assisted surgery suggests Omidria's efficacy, the authors acknowledge these results may be skewed by surgeon preference to avoid using femtosecond laser in patients with preoperative risk factors for intraoperative miosis.

In 2017, Donnenfeld et al.[4▪▪] compared Omidria (483 μmol/l phenylephrine and 89 μmol/l ketorolac in 500 ml BSS) to ketorolac (89 μmol/l ketorolac in 500 ml BSS), phenylephrine (483 μmol/l phenylephrine in 500 ml BSS) and BSS alone on pupil diameter and ocular pain. Intraoperative pupillary mydriasis was significantly larger in the Omidria cohort compared with ketorolac (least mean square difference 0.7 ± 0.1 mm; P < 0.0001) and BSS (least square mean difference 0.9 ± 0.1 mm; P < 0.0001) whereas intraocular pain was significantly reduced for the Omidria cohort compared with phenylephrine (−5.9 ± 2.2; P = 0.009) and BSS (−4.6 ± 2.2; P = 0.042) based on the visual analogue scale (VAS). Finally, significantly fewer patients treated with the study drug (6.1%) had an intraoperative pupil diameter smaller than 6.0 mm at any point during surgery compared with phenylephrine (22.4%, P = 0.0216), ketorolac (34.6%; P = 0.0004) or BSS (47.2%, P < 0.0001). The authors conclude Omidria is a well tolerated and efficacious combination medication that shows superiority to either of its components alone for mydriasis.


Cataract surgeons often use NSAIDs postoperatively to control patient pain, inflammation, and prevent CME, the most prevalent complication affecting postop visual recovery [5▪]. Large retrospective chart reviews performed in 2016 calculated the incidence of clinical postoperative CME between 1.17 and 2.54% [5▪,6,7] in patients with low risk for CME development. They estimated for patients who develop CME, postoperative costs can nearly double with average ophthalmic charges of $10 410 for CME patients compared with $5950 for those without CME [6].

The American Academy of Ophthalmology's Preferred Practice Patterns for Adult Cataract Surgery from 2016 states, ‘There is evidence that NSAIDs, alone or in combination with topical corticosteroids, decrease the likelihood of postoperative CME [8].’ Although some studies have shown benefit for early visual recovery, none have provided convincing Level I evidence. To date, no high-quality meta-analyses, systemic reviews of randomized controlled trials (RCTs) or RCTs with a very low risk of bias have demonstrated a long-term benefit (i.e. 3 months or longer). A Cochrane review by Lim et al.[9] including 34 RCTs published prior to 2016 noted superior BCVA and lower incidence of CME in patients receiving NSAID/steroid combination compared with steroid alone. However, authors deemed this very low certainty evidence as only two cases of CME were reported in the steroid-alone group. Additionally, there were no RCTs published comparing BCVA or CME rates for patients taking NSAIDs alone versus steroids alone at POM 12 visit. In our literature review, no study published between 2016 and 2018 followed low-risk patients beyond 3 months postoperation, and therefore, the long-term benefits of postoperative NSAIDs on visual acuity cannot be commented on.

Lee et al.[10] compared central macular thickness (CMT) measurements by Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, California, USA) between 38 patients taking ketorolac 0.45% twice daily (Acuvail, Allergan, Dublin, Ireland) and 38 patients taking diclofenac 0.1% four times daily (Ofenac, Taejoon, Seoul, Korea) starting 1 day prior to surgery until postoperation week 4 (POW 4). All patients were at low risk for CME development and had uncomplicated cataract surgery with no concurrent use of topical steroid during the study period. CMT measurements of the ketorolac cohort had lower average macular thickness than diclofenac cohort at POM 1 (8.87 ± 9.66 versus 16.36 ± 16.83 μm; P = 0.02) and POM 2 (4.02 ± 10.54 versus 11.92 ± 11.12 μm; P < 0.01). However, no significant differences in BCVA were noted between treatment groups (POM 1, P = 0.78 and POM 2, P = 0.54). The authors concluded that ketorolac, a COX-1 and COX-2 inhibitor, controlled postoperative inflammation better than diclofenac, a selective COX-2 inhibitor. Additionally, ketorolac's less frequent dosing at twice daily likely improves patient compliance compared with diclofenac's dosing at four times daily.

A randomized control trial performed by Stock et al.[11] demonstrated no significant differences in Stratus III OCT (Carl Zeiss Meditec) postoperative CMT values or BCVA at postoperative day (POD) 1, POD 7, and POD 45 when comparing 21 eyes using nepafenac 0.3% daily (Ilevro, Alcon Laboratories Inc., Fort Worth, Texas, USA), 32 eyes using ketorolac 0.5% three times daily (Acular, Syntex, Palo Alto, California, USA), and 24 eyes using propylene glycol three times daily (Systane, Alcon Laboratories Inc.). Study medications were started 2 days preoperation until POD 45, no concurrent steroids were used in the study period and only low-risk patients were included. The authors concluded that all patients tolerated NSAIDs well in the study period. However, because of small sample sizes, no conclusions could be drawn about which regimen was best for preventing postoperative CME. A meta-analysis published in 2017, corroborated these results finding no significant difference in BCVA at POD 1 [weighted mean difference (WMD) 0.084; P = 0.808], POW 1 (WMD −0.008; P = 0.154) and POM 1 (WMD 0.001; P = 0.179) for 262 patients taking Nepafenac and 274 patients taking ketorolac among five randomized control trials [12]. Similarly, no significant differences were found in CMT at POW 1 (WMD −1.455; P = 0.577) and POM 1 (WMD −2.199; P = 0.771) for 141 patients taking Nepafenac and 145 patients taking Ketorolac in two randomized control trials. Of note, medication strength, dosing regimen, and concurrent use of topical steroids was not standardized among studies.

Finally, a randomized control trial performed by Palacio et al.[13] compared CMT measurements between 69 patients taking bromfenac 0.09% three times daily (Zebesten ofteno, Sophia Laboratories SA de CV, Guadalajara, Jalisco, Mexico) and 70 patients taking nepafenac 0.1% three times daily (Nevenac, Alcon Laboratories Inc.). Study medications were started 5 h before surgery and continued until POM 1 visit, with no concurrent topical steroid use and only patients at low risk of CME development were included. Patients in the bromfenac cohort had significantly less CMT thickening at POM 1 compared with the nepafenac cohort patients (252.0 ± 24.9 versus 264.0 ± 34 μm; P = 0.022). No visual acuity data was collected.

In conclusion, while recent studies demonstrated significantly lower CMT measurements in patients who used ketorolac compared with diclofenac, bromfenac compared with nepafenac, and equivalent postoperative CMT measurements for ketorolac compared with nepafenac, no studies demonstrated superior postoperative BCVA or reduction in incidence of CME in patients with low preoperative risk of CME.


Risk factors for the development of postoperative CME include contralateral CME [5▪], posterior capsule rupture [7], epi-retinal membrane (ERM) [5▪,7,14], macular holes [5▪], uveitis [7], retinal vein occlusion [5▪,7], retinal detachment repairs [7], and diabetic retinopathy [5▪,7]. Incidence of clinical CME following cataract surgery in high-risk patients ranges from 4.2% [5▪] to 4.5% [6]. According to the Preferred Practice Patterns from the American Academy of Ophthalmology, ‘The perioperative prophylactic use of NSAIDs for prevention of CME has been advocated for high-risk eyes based on a number of studies.’ However, again there is no level I evidence that visual outcome is improved by the routine use of prophylactic NSAIDs at 3 months or more after cataract surgery in high-risk patients [8].

A retrospective review of 89 731 Kaiser Permanente patient charts concluded that the use of topical NSAIDs reduced the incidence of CME compared with those not receiving topical NSAIDs for all patients (1.3 versus 1.7%; P < 0.001), patients without diabetes (1.1 versus 1.5%; P < 0.001), and diabetics without retinopathy (0.7 versus 1.1%; P = 0.006) [15▪]. Patients with diabetic retinopathy who used topical NSAIDs, however, did not show a significant reduction in incidence of CME compared with patients with diabetic retinopathy who did not use topical NSAIDs (3.4 versus 3.2%; P = 0.650). The authors concluded that the incidence of CME in the absence of perioperative NSAIDs is nominal for a majority of patients, and therefore, prophylactic NSAIDs have limited utility.

In 2018, Alnagdy et al.[16] assessed the impact of topical nepafenac 0.1% three times daily, ketorolac tromethamine 0.4% four times daily, and placebo four times daily on patients diagnosed with diabetes without diabetic retinopathy. The study medications were started two days prior to surgery, ended 2 months after surgery and were co-administered with a topical antibiotic-steroid four times daily for 3 weeks. At POM 3 BCVA was significantly better in both NSAID groups compared with placebo (P = 0.04). Furthermore, patients in the NSAID cohort had significantly less CMT thickening by Topcon 3D 1000OCT (Hasunuma-cho, Itabashi-Ku, Tokyo, Japan) measurements at POM 1 (234.5 ± 26.16 versus 255.7 ± 33.2 μm; P = 0.08), POM 2 (235.43 ± 24.21 versus 260.41 ± 55.8 μm; P = 0.027) and POM 3 (231.7 ± 21.5 versus 267.5 μm; P = 0.004) compared with the placebo cohort. No statistically significant differences in BCVA or CMT were found between nepafenac (0.1%) and ketorolac (0.4%).

When evaluating patients with nonproliferative diabetic retinopathy (NPDR) without prior diabetic macular edema (DME), Singh et al.[17] demonstrated a significant reduction in incidence of CME for patients taking nepafenac 0.3% daily compared with vehicle (4.1 versus 15.19%; P < 0.001). Additionally, significantly more patients achieved BCVA gain of greater than 15 letters from preoperative baseline in the nepafenac cohort compared with vehicle (55.4 versus 46.7%, P = 0.003). Although the authors demonstrated a reduction of the incidence of CME with nepafenac use in patients with diabetic retinopathy, the severity of NPDR and response to NSAID therapy was not stratified. A similar study by Sarfraz et al.[18] compared nepafenac 0.3% four times daily for 3 months with topical prednisolone 0.1% every 4 h for 2 weeks (Allergan) to topical prednisolone alone in patients with NPDR without DME. 3.3% of patients receiving nepafenac developed CME on SD-OCT (3D OCT-1000, Topcon Co, Tokyo Japan) in the study period compared with 23.3% of patients receiving steroids alone. Similarly, the nepafenac cohort demonstrated significantly less CMT thickening at POM 3 compared with steroid alone (2.33 ± 10.45 versus 12.23 ± 12.40 μm; P = 0.023). Data was not stratified by diabetic retinopathy severity.


The optimal regimen for preventing postoperative pain and discomfort was widely studied between 2016 and 2018. A Cochrane review involving 48 randomized control trials by Juthani et al.[19▪] attempted to determine whether monotherapy with NSAIDs or corticosteroids is superior for controlling postoperation inflammation. Reviewers found moderate certainty evidence that there was no difference in cell grading between treatment groups [mean difference 0.60; 95% confidence interval (CI) −2.19 to 0.99]; however, there was low certainty evidence that NSAIDs alone were superior to corticosteroids alone for controlling flare (mean difference −13.74, 95% CI −21.45 to −6.0).

When utilizing corticosteroids in combination with various NSAIDs, a meta-analysis by Zhao et al.[12] demonstrated nepafenac significantly reduced postoperative ocular discomfort and hypreremia compared with ketorolac [relative risk (RR) = 0.589, 95% CI 0.436–0.794; P = 0.001 and RR = 0.253, 95% CI 0.115–0.557; P = 0.001 respectively]. No significant differences in anterior chamber inflammation grading were found between medications. Additionally, when comparing bromfenac 0.09% with nepafenac 0.3% no significant differences in the presence of pain, chemosis, photophobia, flare, hyperemia, cellularity or corneal edema were observed at 30 days following cataract surgery [13].


Although postoperative pain and inflammation can cause significant discomfort for cataract surgery patients, Chatziralli et al.[20] analyzed postoperation visual quality of life using VFQ25 questionnaires. No significant differences in the composite scores were found for 70 patients taking Ketorolac 0.5% three times daily compared with 68 patients taking placebo at POD 7, POD 28, and POD 42 (P = 0.833, P = 0.360, and 0.467, respectively). The authors argued that NSAIDs do not improve quality of vision in low-risk patients following cataract surgery.


The best use of perioperative NSAIDs continues to be a source of debate in the ophthalmologic literature. Reviewing articles published between 1 January 2016 and 30 June 2018 revealed new developments at every stage of cataract surgery. Perioperative NSAIDs achieve better intraoperative mydrasis, lower rates of postoperative CME, and improve patient comfort following surgery. Further high-quality studies are needed to establish Level I evidence on the long-term effects of perioperative NSAIDs on visual acuity and CME rates in low-risk and high-risk patients.



Financial support and sponsorship

Ronald M. Burde, M.D. Residency Education Fund.

Conflicts of interest

There are no conflicts of interest.


Papers of particular interest, published within the annual period of review, have been highlighted as:

  • ▪ of special interest
  • ▪▪ of outstanding interest


1▪. Katsev DA, Katsev CC, Pinnow J, Lockhart CM. Intracameral ketorolac concentration at the beginning and end of cataract surgery following preoperative topical ketorolac administration. Clin Ophthalmol 2017; 11:1897–1901.

Clinical study on patients establishing the washout effect of intraoperative irrigation on preoperative NSAIDs and defining the need for intracameral NSAIDs to mitigate operation-induced inflammation.

2. Tomislav AM, Matas JA, Vukojevic K, et al. Effect of preoperative topical diclofenac on intraocular interleukin-12 concentration and macular edema after cataract surgery in patients with diabetic retinopathy: a randomized controlled trial. Croat Med J 2017; 58:49–55.
3. Bucci FA Jr, Michalek B, Fluet AT. Comparison of the frequency of use of a pupil expansion device with and without an intracameral phenylephrine and ketorolac injection 1%/0.3% at the time of routine cataract surgery. Clin Ophthalmol 2017; 11:1039–1043.
4▪▪. Donnenfeld ED, Whitaker S, Jackson MA, Wittpen J. Intracameral ketorolac and phenylephrine effect on intraoperative pupil diameter and postoperative pain in cataract surgery. J Cataract Refract Surg 2017; 43:597–605.

Randomized control trial, which demonstrates the superiority of intracameral 1%/0.3% phenylephrine/ketorolac combination compared with each component alone in reducing the need for intraoperative pupillary expansion devices. Authors’ sub-group analysis shows benefits for patients taking alpha-1 blocking medication at high risk of IFIS but is inconclusive for femtosecond laser-assisted cataract surgery patients.

5▪. McCafferty S, Harris A, Kew C, et al. Pseudophakic cystoid macular edema prevention and risk factors; prospective study with adjunctive once daily topical nepafenac 0.3% versus placebo. BMC Ophthalmol 2017; 17:16.

A prospective randomized control trial showing nepafenac prevents CME in high-risk patients but does not affect CME rates in low-risk populations. Authors argue the use of NSAIDs should be reserved for patients at risk of developing postoperative CME.

6. Schmier JK, Covert DW, Hulme-Lowe CK, et al. Treatment costs of cystoid macular edema among patients following cataract surgery. Clin Ophthalmol 2016; 10:477–483.
7. Chu CJ, Johnston RL, Buscombe C, et al. Risk factors and incidence of macular edema after cataract surgery a database study of 81984 eyes. Ophthalmology 2016; 123:316–323.
8. Olson RJ, Braga-Mele R, Chen SH, et al. Cataract in the adult eye preferred practice pattern ®. Ophthalmology 2017; 124:P1–P119.
9. Lim BX, Lim CH, Lim DK, et al. Prophylactic nonsteroidal anti-inflammatory drugs for the prevention of macular oedema after cataract surgery. Cochrane Database Syst Rev 2016; 11:CD006683.
10. Lee TH, Choi W, J YS, Yoon KC. Comparison of ketorolac 0.45% versus diclofenac 0.1% for macular thickness and volume after uncomplicated cataract surgery. Acta Ophthalmol 2016; 94:e177–e182.
11. Stock RA, Galvan DK, Godoy R, Bonamigo EL. Comparison of macular thickness by optical coherence tomography measurements after uneventful phacoemulsification using ketorolac tromethamine, nepafenac, versusa control group, preoperatively and postoperatively. Clin Ophthalmol 2018; 12:607–611.
12. Zhao X, Xia S, Wang E, Chen Y. Comparison of the efficacy and patients’ tolerability of nepafenac and ketorolac in the treatment of ocular inflammation following cataract surgery: a meta-analysis of randomized controlled trials. PLoS One 2017; 12:e0173254.
13. Palacio C, Ortega LFD, Bustos FR, et al. Bromfenac 0.09% bioavailability in aqueous humor, prophylactic effect on cystoid macular edema, and clinical signs of ocular inflammation after phacoemulsification in a Mexican population. Clin Ophthalmol 2016; 10:233–237.
14. Kim S, Kim MK, Wee WR. Additive effect of oral steroid with topical nonsteroidal anti-inflammatory drug for preventing cystoid macular edema after cataract surgery in patients with epiretinal membrane. Korean J Ophthalmol 2017; 31:394–401.
15▪. Modjtahedi BS, Paschal JF, Batech M, et al. Perioperative topical nonsteroidal anti-inflammatory drugs for macular edema prophylaxis following cataract surgery. Am J Ophthalmol 2017; 176:174–182.

Large retrospective review of 89 731 patients following cataract surgery establishing NSAIDs help prevent clinical CME in patients without diabetes and diabetic patients without retinopathy, however, patients with diabetic retinopathy showed no reduction in CME rates with NSAID use.

16. Alnagdy AA, Abouelkheir HY, El-Khouly SE, Tarshouby SM. Impact of topical nonsteroidal anti-inflammatory drugs in prevention of macular edema following cataract surgery in diabetic patients. Int J Ophthalmol 2018; 11:616–622.
17. Singh RP, Lehmann R, Martel J, et al. Nepafenac 0.3% after cataract surgery in patients with diabetic retinopathy results of 2 randomized phase 3 studies. Ophthalmology 2017; 124:776–785.
18. Sarfraz MH, Haq RI, Mehboob MA. Effect of topical nepafenac in prevention of macular edema after cataract surgery in patients with nonproliferative diabetic retinopathy. Pak J Med Sci 2017; 33:210–214.
19▪. Juthani VV, Clearfield E, Chuck RS. Nonsteroidal anti-inflammatory drugs versus corticosteroids for controlling inflammation after uncomplicated cataract surgery. Cochrane Database Syst Rev 2017; 7:CD010516.

First large-scale Cochrane review of 48 randomized control trials comparing NSAID alone, corticosteroid alone and NSAID/corticosteroid combination for the prevention of CME and controlling postoperative inflammation.

20. Chatziralli IP, Sergentanis TN, Parikakis EA, et al. The impact of nonsteroidal anti-inflammatory agents after phacoemulsification on quality of life: a randomized study. Ophthalmol Ther 2017; 6:133–140.

intraoperative mydriasis; nonsteroidal anti-inflammatory drug; postoperative cystoid macular edema

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.