- Email: [email protected]
Prednisolone and flurbiprofen drops to maintain mydriasis during phacoemulsification cataract surgery
Prednisolone and flurbiprofen drops to maintain mydriasis during phacoemulsification cataract surgery M. Yusuf Shaikh, FRCS, J. Simon Mars, FRCOphth, Chris J. Heaven, FRCS Purpose: To compare the effects on the pupil of pretreatment with prednisolone 1% eyedrops, flurbiprofen 0.03% eyedrops, and sodium chloride 0.9% eyedrops in patients having phacoemulsification cataract surgery. Setting: Christopher Home Eye Unit, Royal Albert Edward Infirmary, Wigan, United Kingdom. Methods: One hundred thirty-six white patients having phacoemulsification with no history of eye disease, antiinflammatory medication, or diabetes were enrolled in this prospective randomized double-blind study. Patients were allocated to receive flurbiprofen 0.03%, prednisolone 1%, or sodium chloride 0.9% (placebo) eyedrops at 30-minute intervals beginning 2 hours before surgery. All patients received cyclopentolate 1% and phenylephrine 2.5% drops preoperatively and epinephrine 1:106 intraoperatively in the intraocular irrigating solution. The pupil diameter was measured from a television screen at the preset magnification of the operating microscope immediately before the incision, at the end of nucleus emulsification, at the end of irrigation/aspiration (I/A), and at the end of surgery. One hundred twelve patients with a baseline pupil diameter of at least 6.0 mm who had uneventful phacoemulsification were analyzed. Results: The baseline pupil diameters were statistically similar. An analysis of variance did not reveal an overall significant mydriasis-maintaining effect of prednisolone or flurbiprofen compared to the placebo at any stage of surgery (P ⫽ .16 to .37). The 95% confidence interval showed flurbiprofen to be superior to prednisolone at the end of surgery. According to the paired t test, the reduction from the baseline at the end of I/A in the prednisolone group (P ⫽ .051) and at the end of surgery in the flurbiprofen group (P ⫽ .22) was not significant. Conclusion: In the presence of epinephrine in the intraocular irrigating solution, both prednisolone 1% and flurbiprofen 0.03% failed to maintain mydriasis at the crucial steps of nuclear emulsification and cortical I/A. J Cataract Refract Surg 2003; 29:2372–2377 2003 ASCRS and ESCRS
T
he miosis that occurs during cataract surgery is in part mediated by prostaglandins (PGs).1 Preoperative treatment with nonsteroidal antiinflammatory drugs (NSAIDs) has been shown to be effective in maintaining mydriasis during cataract surgery.2 The mechanism of their action is dependent on their ability to cause cycloAccepted for publication January 27, 2003. Reprint requests to J.S. Mars, Christopher Home Eye Unit, Royal Albert Edward Infirmary, Wigan Lane, Wigan, WN1 2NN, United Kingdom. E-mail: [email protected]. 2003 ASCRS and ESCRS Published by Elsevier Inc.
oxygenase inhibition and thereby inhibit the production of PGs in response to surgical trauma (Figure 1).3 Corticosteroids block the release of arachidonic acid, which is a precursor of the synthesis of PGs and other chemical mediators such as leukotrienes.4 Unlike nonsteroidal agents, corticosteroid eyedrops have not been extensively studied for their antimiosis effect. Limited effectiveness of prednisolone 1% eyedrops in maintaining mydriasis during conventional large-incision extracapsular cataract extraction (ECCE) has been documented.5 0886-3350/03/$–see front matter doi:10.1016/S0886-3350(03)00137-8
PREDNISOLONE AND FLURBIPROFEN FOR MYDRIASIS DURING PHACOEMULSIFICATION
Figure 1. (Shaikh) Schematic of the arachidonic acid pathway.
It remains to be ascertained whether a significant miosis-inhibiting effect can be obtained in modern small-incision phacoemulsification surgery. We therefore devised a randomized prospective placebo-controlled study to assess the miosis-inhibiting property of topical prednisolone 1% eyedrops and flurbiprofen 0.03% eyedrops. We chose flurbiprofen as a nonsteroidal agent for comparison as it is widely used and has been approved by the U.S. Food and Drug Administration (FDA) for use in maintaining mydriasis during cataract surgery.
Patients and Methods White patients of either sex aged 55 years or older with age-related cataract requiring phacoemulsification surgery were enrolled in this randomized double-masked placebocontrolled comparative study. An institutional ethical committee approved the study, and patients provided informed consent before their enrollment. Patients were excluded if they had a history of eye disease (glaucoma, uveitis), ocular trauma, or ocular surgery. Use of topical or systemic steroids or nonsteroidal agents within the 2 weeks before surgery, a suggestion in the history of hypersensitivity to any study material, and diabetes mellitus were other exclusion criteria. The eligible patients were randomly allocated to 1 of 3 groups according to a computer-generated randomization schedule: sodium chloride 0.9% eyedrops (placebo), prednisolone acetate 1% eyedrops (Pred Forte威), or flurbiprofen sodium 0.03% eyedrops (Ocufen威). It was estimated that 108 patients (36 in each group) would provide a sample size with 95% power to detect a 0.4 mm difference in pupil diameters. Each patient received a drop of his or her allocated treatment at 3 30-minute intervals starting 2 hours before
surgery. The routine dilation regimen in all 3 groups consisted of 3 applications of 1 drop each of cyclopentolate 1% and phenylephrine 2.5% at 30-minute intervals. All patients received an anterior retrobulbar injection of a mixture of equal amounts of lidocaine 2% and bupivacaine 0.75%. The phacoemulsification was performed according to standard operating procedure to ensure uniformity of technique. All patients had clear corneal sutureless phacoemulsification cataract extraction with implantation of a foldable plate-haptic silicone posterior chamber intraocular lens (IOL). The viscoelastic material was sodium hyaluronate 1% (Healon威) and the irrigating solution, a balanced salt solution with adrenaline 1:106. Patients were excluded from analysis if the horizontal pupil diameter at the start of surgery was smaller than 6.0 mm or if there was posterior capsule rupture, zonule dehiscence, direct iris trauma, or intraocular bleeding complicating the surgical procedure. All surgeries were videotaped. The horizontal pupil diameter was measured from the frozen frames of the replay on the television screen. The pupil diameter was measured during surgery at 4 points: (1) before the corneal incision; (2) after complete removal of the nucleus by phacoemulsification; (3) after removal of all cortical soft lens matter; (4) at the end of surgery before subconjunctival injection of an antibiotic. All measurements were taken when the eye had an inflated anterior chamber without viscoelastic material. The preset standard magnification of the operating microscope was ensured at the above 4 points. The magnification was done previously by placing a millimeter scale under the microscope. The measurements were converted to the actual pupil size. The patient, surgeon, and investigator measuring the pupil diameters were masked to the treatment used. Demographic characteristics were summarized using frequency counts. Time interval data were tested using the Shapiro-Wilks test and univariate summary statistics. The distributions were approximately normal. The data were summarized using mean, standard deviation, number of observations, and the minimum and maximum. The changes in pupil diameter from the baseline values before the corneal incision were analyzed using a 1-way analysis of variance (ANOVA) to allow for treatment group effects. Both active treatment groups were compared with the placebo group. The 95% confidence interval (CI) of the difference between flurbiprofen and prednisolone was calculated and compared to the equivalence interval of –0.2 to ⫹0.2 mm. Within the group, the changes in pupil diameter from the baseline value at each stage of the procedure were analyzed using the paired t test.
Results Of 136 enrolled patients, 112 qualified for final analysis. The groups were well matched in demographic characteristics and surgical data (Table 1).
J CATARACT REFRACT SURG—VOL 29, DECEMBER 2003
2373
PREDNISOLONE AND FLURBIPROFEN FOR MYDRIASIS DURING PHACOEMULSIFICATION
The mean pupil diameters before the corneal incision, at the end of the phacoemulsification, at the end of irrigation/aspiration (I/A), and at the end of surgery
Figure 2. (Shaikh) Mean change in pupil diameter from baseline (End Emulsification ⫽ end of nuclear phacoemulsification; End A/I ⫽ end of aspiration and irrigation; End Surgery ⫽ end of the procedure).
Table 1. Demographic characteristics and surgical data of patients. Parameter
Placebo
Patients, n
36
38
38
73.3 ⫾ 7.6
76.4 ⫾ 8.1
74.2 ⫾ 9.2
14
20
18
24
18
20
Iris, n Light
32
35
33
Dark
4
3
5
2.2 ⫾ 0.82
2.4 ⫾ 0.86
2.2 ⫾ 0.92
Mean age, y Sex, n Male Female
Mean phaco time Mean phaco %
19.5 ⫾ 5.2
Flurbiprofen Prednisolone
21.6 ⫾ 6.5
20.5 ⫾ 6.6
Means ⫾ SD
are shown in Table 2. Although the mean pupil diameter at baseline tended to be larger in the flurbiprofen and prednisolone groups than in the placebo group, the differences between groups were not statistically significant (P ⫽ .65). The mean pupil size at the subsequent stages of the surgical procedure was invariably smaller than the baseline value in all 3 groups (Table 3 and Figure 2). The paired t test showed that this decrease from baseline was significant at the end of phacoemulsification in all 3 groups. However, at the end of I/A in the prednisolone group (P ⫽ .51) and at the end of surgery in the flurbiprofen group (P ⫽ .22), the reduction was not statistically significant. Intergroup comparison with the 1-way ANOVA showed that overall, there was no statistically significant difference between the 3 groups at any stage of surgery (P ⫽ .21 at the end of emulsification, .37 at the end of I/A, and .16 at the end of the surgery). In the placebo group, the mean decrease in pupil diameter from the baseline ranged from 0.28 to 0.33 mm at various stages of surgery (Table 3). The mean decreases in the other groups were smaller, ranging from 0.08 to 0.17 mm in the flurbiprofen group and 0.16 to 0.22 mm in the prednisolone group. Table 4 shows the 95% CI of the difference between flurbiprofen and prednisolone at each stage of surgery. The latter CI reflects a change in pupil diameter of 0.4 mm, which was considered the smallest clinically significant difference that could affect the conduct of the procedure. The 2 treatment groups were equivalent at all stages except at the end of surgery, when flurbiprofen provided a significantly better mydriasis-maintaining effect than prednisolone. The proportion of patients in each treatment group showing a decrease of 0.4 mm or more from the baseline pupil diameter is shown in Figure 3. The differences between the groups were not statistically significant
Table 2. Mean pupil diameters at various stages of surgery. Mean Diameter (mm) ⫾ SD Group
Before Corneal Incision
End of Emulsification
End of I/A
End of Surgery
Placebo
7.89 ⫾ 0.833
7.56 ⫾ 0.876
7.60 ⫾ 0.844
7.61 ⫾ 0.895
Flurbiprofen
8.08 ⫾ 1.002
7.91 ⫾ 0.997
7.90 ⫾ 1.010
8.00 ⫾ 1.054
Prednisolone
7.92 ⫾ 1.046
7.70 ⫾ 0.991
7.76 ⫾ 1.023
7.73 ⫾ 1.019
I/A ⫽ irrigation/aspiration
2374
J CATARACT REFRACT SURG—VOL 29, DECEMBER 2003
PREDNISOLONE AND FLURBIPROFEN FOR MYDRIASIS DURING PHACOEMULSIFICATION
Table 3. Change in pupil diameter (mm) from baseline. End of Emulsification
End of I/A
End of Surgery
Group
Mean ⫾ SD
P Value
Mean ⫾ SD
P Value
Mean ⫾ SD
P Value
Placebo
⫺0.33 ⫾ 0.366
.0001
⫺0.29 ⫾ 0.397
.0001
⫺0.28 ⫾ 0.462
.0008
Flurbiprofen
⫺0.17 ⫾ 0.301
.0011
⫺0.18 ⫾ 0.337
.0020
⫺0.08 ⫾ 0.397
.027
Prednisolone
⫺0.22 ⫾ 0.483
.0087
⫺0.16 ⫾ 0.483
.051
⫺0.18 ⫾ 0.493
.22
I/A ⫽ irrigation/aspiration
Table 4. The 95% CIs for the difference of means between the fluorbiprofen and prednisolone groups. Stage of Surgery
95% CI
End of emulsification
⫺0.081, ⫹0.170
End of I/A
⫺0.155, ⫹0.108
End of surgery
⫺0.038, ⫹0.0250
CI ⫽ confidence interval; I/A ⫽ irrigation/aspiration
(P ⫽ .32 at the end of emulsification, .29 at the end of I/A, and .33 at the end of surgery). In all groups, miosis of 1.0 mm or more was observed in no more than 5% of patients at any stage. Figure 4 shows the frequency distribution of the different magnitudes of pupil change in the 3 groups at various stages.
Figure 3. (Shaikh) Proportion of patients with miosis 0.4 mm or more from baseline (End Phaco ⫽ end of nuclear phacoemulsification; End A/I ⫽ end of aspiration and irrigation; End surgery ⫽ end of the procedure).
Discussion Pretreatment with various NSAIDs reduces intraoperative miosis during cataract surgery.2 Preoperative treatment is key because once PGs are released, a topical NSAID will not block their effect on the iris. Preoperative treatment with prednisolone acetate also reduces surgical miosis to a limited extent during ECCE.5 In the present study, ANOVA analysis showed that both flurbiprofen and prednisolone failed to have a statistically significant mydriasis-maintaining effect in comparison to the placebo at any stage of surgery. That the patients in all 3 groups in our study had epinephrine in the intraocular irrigating solution could have dictated this outcome. Intraoperative epinephrine has been shown to be the most effective single factor in reducing surgical miosis.6 Our results agree with those of Gimbel, 6 who reports no significant differences among epinephrinetreated eyes with or without pretreatment with NSAIDs.6 Inclusion of treatment groups in which intra-
Figure 4. (Shaikh) Frequency distribution of pupil change from baseline. The y-axis denotes various treatment groups using the following abbreviations: EE is the end of emulsification, EA is the end of aspiration irrigation, and ES is the end of surgery. The x-axis denotes the change in mean pupil diameter using the following abbreviations: ⬎⫽ is more than or equal to, ⬍⫽ is less than or equal to, – is a reduction in diameter, and ⫹ is an increase in diameter.
J CATARACT REFRACT SURG—VOL 29, DECEMBER 2003
2375
PREDNISOLONE AND FLURBIPROFEN FOR MYDRIASIS DURING PHACOEMULSIFICATION
operative epinephrine was not used could have clarified the effects of the drugs used in isolation. However, many phacosurgery units routinely use intraocular adrenaline as a mydriatic agent. This might also explain why the routine use of anti-PGs is not universally accepted. With regard to prednisolone, our results agree with those in a study of ECCE in which prednisolone produced a limited beneficial effect that did not reach statistical significance.5 Our findings reinforce the belief that surgical miosis may be mediated in part by as yet unidentified factors independent of PG pharmacodynamics.7 Also, the differences in dosing regimens, surgical techniques, and patient populations affect the outcomes measure.6 In an intragroup comparison by a paired t test at the crucial stages of cortical cleanup and IOL implantation, both prednisolone and flurbiprofen failed to show a significant antimiotic effect. However, there was an indication at the end of I/A in the prednisolone group of a miosis-inhibiting effect that was statistically significant. This effect could have been spurious as it was not substantiated by the CIs of the difference in the means between the 2 groups. In the flurbiprofen group, significant maintenance of mydriasis at the end of the surgery was suggested by the paired t test and CI data. This observation could be the manifestation of the wellknown phenomenon of synergism between NSAIDs and epinephrine.6 The incidence of miosis of more than 0.4 mm in the flurbiprofen group was about half that in the placebo group. Although the differences between the groups were not statistically significant, this is a clinically significant phenomenon that may be further substantiated by a study repeated on a larger sample. In eyes that are dilated preoperatively with phenylephrine and cyclopentolate and have received no pretreatment with the nonsteroidal drops, Gimbel6 has shown that the mean miosis during phacoemulsification when intraoperative epinephrine was used was 0.2 mm ⫾ 0.059 (SEM). Therefore, the CIs chosen for the equivalence comparison in our study were –0.2 to ⫹0.2 mm. These CIs reflect a change in pupil diameter of 0.4 mm, which we considered the smallest clinically significant difference that could affect the conduct of the procedure. The threshold of pupil inequality clinically detectable to the naked eye is also 0.4 mm. Using the aforementioned equivalence interval, the effect of prednisolone and flurbiprofen was equivalent throughout 2376
the procedure except at the end of surgery, when flurbiprofen gave significantly better effects. Because prednisolone blocks the arachidonic acid metabolism cascade at a higher level, both the cyclooxygenase and lipooxygenase pathways are blocked. The more specific blockage of the cyclooxygenase pathway by the nonsteroidal agent could, in addition to the possible synergism with epinephrine, explain this result. Moreover, the specific mechanism of action of cyclooxygenase blockers such as flurbiprofen may have a variety of biological effects that cannot be satisfactorily explained by inhibition of PG synthesis alone.8 We chose flurbiprofen as a nonsteroidal agent in our study as it has FDA approval for specific use in maintaining mydriasis during cataract surgery. However, newer and more potent NSAIDs such as diclofenac and ketorolac, which are gaining wider acceptance, may have more favorable effects on the pupil and additional benefits with regard to cystoid macular edema (CME).9,10 Prednisolone in combination with ketorolac was shown to have a better effect than either agent alone in the treatment of CME.11 It would be interesting to see whether the similar synergistic antimiotic effect of the combination can be obtained. In our study, uniform and minimal surgical trauma was ensured by excluding cases with a baseline pupil diameter smaller than 6.0 mm. The greater release of inflammatory mediators anticipated in small-pupil surgery may be more susceptible to modification by therapeutic agents. There are also cases in which maintaining mydriasis would benefit the surgeon. Retrobulbar anesthesia was used in our study to achieve uniform anesthesia and akinesia in all patients. Under topical anesthesia, different neuropharmacology considerations may be involved and, therefore, interesting results could be expected in a study using topical anesthesia. An important limitation of our study was that the pretreatment began only 2 hours before surgery. Treatment starting up to 3 days preoperatively is suggested to optimize the antiinflammatory effects of diclofenac.12 Many anterior segment subspecialists start treatment with an NSAID as early as 3 days preoperatively to achieve maximum prophylaxis against intraoperative miosis and postoperative CME. Further study using such a regimen may yield additional and potentially valuable information.
J CATARACT REFRACT SURG—VOL 29, DECEMBER 2003
PREDNISOLONE AND FLURBIPROFEN FOR MYDRIASIS DURING PHACOEMULSIFICATION
References 1. Duffin RM, Camras CB, Gardner SK, Pettit TH. Inhibitors of surgically induced miosis. Ophthalmology 1982; 89:966–997; discussion by SM Poder, 978–979 2. Nichols J, Synder RW. Topical nonsteroidal anti-inflammatory agents in ophthalmology. Curr Opin Ophthalmol 1998; 9(4):40–44 3. Podos SM. Prostaglandins, nonsteroidal anti-inflammatory agents and eye disease. Trans Am Ophthalmol Soc 1976; 74:637–660 4. Kulkarni PS. Steroids in ocular therapy. In: Zimmerman TJ, ed, Textbook of Ocular Pharmacology. Philadelphia, PA, Lippincott-Raven, 1997; 63–67 5. Dube´ P, Boisjoly HM, Bazin R, et al. Comparison of prednisolone acetate and indomethacin for maintaining mydriasis during cataract surgery. Can J Ophthalmol 1990; 25:234–238 6. Gimbel HV. The effect of treatment with topical nonsteroidal anti-inflammatory drugs with and without intraoperative epinephrine on the maintenance of mydriasis during cataract surgery. Ophthalmology 1989; 96:585– 588 7. Flach AJ. Nonsteroidal anti-inflammatory drugs in ophthalmology. Int Ophthalmol Clin 1993; 33(4):1–7 8. To KW, Abelson MB, Neufield A. Nonsteroidal antiinflammatory drugs. In: Albert DM, Jakobiec FA, eds, Principles and Practice of Ophthalmology; Basic Sci-
9.
10.
11.
12.
ences. Philadelphia, PA, WB Saunders, 1994; vol 6, 1022–1027 Diestelhorst M, Schmidl B, Konen W, et al. Efficacy and tolerance of diclofenac sodium 0.1%, flurbiprofen 0.03%, and indomethacin 1.0% in controlling postoperative inflammation. J Cataract Refract Surg 1996; 22: 788–793 Solomon KD, Turkalj JW, Whiteside SB, et al. Topical 0.5% ketorolac vs 0.03% flurbiprofen in inhibition of miosis during cataract surgery. Arch Ophthalmol 1997; 115:1119–1122 Heier JS, Topping TM, Baumann W, et al. Ketorolac versus prednisolone versus combination therapy in the treatment of acute pseudophakic cystoid macular edema. Ophthalmology 2000; 107:2034–2038; discussion by AJ Flach, 2039 Roberts CW. Pretreatment with topical diclofenac sodium to decrease postoperative inflammation. Ophthalmology 1996; 103:636–639
From Christopher Home Eye Unit, Royal Albert Edward Infirmary, Wigan, United Kingdom. Presented at a meeting of the Saudi Ophthalmological Society, Riyadh, Saudi Arabia, March 2002. None of the authors has a financial or proprietary interest in any material or method mentioned.
J CATARACT REFRACT SURG—VOL 29, DECEMBER 2003
2377
T
he miosis that occurs during cataract surgery is in part mediated by prostaglandins (PGs).1 Preoperative treatment with nonsteroidal antiinflammatory drugs (NSAIDs) has been shown to be effective in maintaining mydriasis during cataract surgery.2 The mechanism of their action is dependent on their ability to cause cycloAccepted for publication January 27, 2003. Reprint requests to J.S. Mars, Christopher Home Eye Unit, Royal Albert Edward Infirmary, Wigan Lane, Wigan, WN1 2NN, United Kingdom. E-mail: [email protected]. 2003 ASCRS and ESCRS Published by Elsevier Inc.
oxygenase inhibition and thereby inhibit the production of PGs in response to surgical trauma (Figure 1).3 Corticosteroids block the release of arachidonic acid, which is a precursor of the synthesis of PGs and other chemical mediators such as leukotrienes.4 Unlike nonsteroidal agents, corticosteroid eyedrops have not been extensively studied for their antimiosis effect. Limited effectiveness of prednisolone 1% eyedrops in maintaining mydriasis during conventional large-incision extracapsular cataract extraction (ECCE) has been documented.5 0886-3350/03/$–see front matter doi:10.1016/S0886-3350(03)00137-8
PREDNISOLONE AND FLURBIPROFEN FOR MYDRIASIS DURING PHACOEMULSIFICATION
Figure 1. (Shaikh) Schematic of the arachidonic acid pathway.
It remains to be ascertained whether a significant miosis-inhibiting effect can be obtained in modern small-incision phacoemulsification surgery. We therefore devised a randomized prospective placebo-controlled study to assess the miosis-inhibiting property of topical prednisolone 1% eyedrops and flurbiprofen 0.03% eyedrops. We chose flurbiprofen as a nonsteroidal agent for comparison as it is widely used and has been approved by the U.S. Food and Drug Administration (FDA) for use in maintaining mydriasis during cataract surgery.
Patients and Methods White patients of either sex aged 55 years or older with age-related cataract requiring phacoemulsification surgery were enrolled in this randomized double-masked placebocontrolled comparative study. An institutional ethical committee approved the study, and patients provided informed consent before their enrollment. Patients were excluded if they had a history of eye disease (glaucoma, uveitis), ocular trauma, or ocular surgery. Use of topical or systemic steroids or nonsteroidal agents within the 2 weeks before surgery, a suggestion in the history of hypersensitivity to any study material, and diabetes mellitus were other exclusion criteria. The eligible patients were randomly allocated to 1 of 3 groups according to a computer-generated randomization schedule: sodium chloride 0.9% eyedrops (placebo), prednisolone acetate 1% eyedrops (Pred Forte威), or flurbiprofen sodium 0.03% eyedrops (Ocufen威). It was estimated that 108 patients (36 in each group) would provide a sample size with 95% power to detect a 0.4 mm difference in pupil diameters. Each patient received a drop of his or her allocated treatment at 3 30-minute intervals starting 2 hours before
surgery. The routine dilation regimen in all 3 groups consisted of 3 applications of 1 drop each of cyclopentolate 1% and phenylephrine 2.5% at 30-minute intervals. All patients received an anterior retrobulbar injection of a mixture of equal amounts of lidocaine 2% and bupivacaine 0.75%. The phacoemulsification was performed according to standard operating procedure to ensure uniformity of technique. All patients had clear corneal sutureless phacoemulsification cataract extraction with implantation of a foldable plate-haptic silicone posterior chamber intraocular lens (IOL). The viscoelastic material was sodium hyaluronate 1% (Healon威) and the irrigating solution, a balanced salt solution with adrenaline 1:106. Patients were excluded from analysis if the horizontal pupil diameter at the start of surgery was smaller than 6.0 mm or if there was posterior capsule rupture, zonule dehiscence, direct iris trauma, or intraocular bleeding complicating the surgical procedure. All surgeries were videotaped. The horizontal pupil diameter was measured from the frozen frames of the replay on the television screen. The pupil diameter was measured during surgery at 4 points: (1) before the corneal incision; (2) after complete removal of the nucleus by phacoemulsification; (3) after removal of all cortical soft lens matter; (4) at the end of surgery before subconjunctival injection of an antibiotic. All measurements were taken when the eye had an inflated anterior chamber without viscoelastic material. The preset standard magnification of the operating microscope was ensured at the above 4 points. The magnification was done previously by placing a millimeter scale under the microscope. The measurements were converted to the actual pupil size. The patient, surgeon, and investigator measuring the pupil diameters were masked to the treatment used. Demographic characteristics were summarized using frequency counts. Time interval data were tested using the Shapiro-Wilks test and univariate summary statistics. The distributions were approximately normal. The data were summarized using mean, standard deviation, number of observations, and the minimum and maximum. The changes in pupil diameter from the baseline values before the corneal incision were analyzed using a 1-way analysis of variance (ANOVA) to allow for treatment group effects. Both active treatment groups were compared with the placebo group. The 95% confidence interval (CI) of the difference between flurbiprofen and prednisolone was calculated and compared to the equivalence interval of –0.2 to ⫹0.2 mm. Within the group, the changes in pupil diameter from the baseline value at each stage of the procedure were analyzed using the paired t test.
Results Of 136 enrolled patients, 112 qualified for final analysis. The groups were well matched in demographic characteristics and surgical data (Table 1).
J CATARACT REFRACT SURG—VOL 29, DECEMBER 2003
2373
PREDNISOLONE AND FLURBIPROFEN FOR MYDRIASIS DURING PHACOEMULSIFICATION
The mean pupil diameters before the corneal incision, at the end of the phacoemulsification, at the end of irrigation/aspiration (I/A), and at the end of surgery
Figure 2. (Shaikh) Mean change in pupil diameter from baseline (End Emulsification ⫽ end of nuclear phacoemulsification; End A/I ⫽ end of aspiration and irrigation; End Surgery ⫽ end of the procedure).
Table 1. Demographic characteristics and surgical data of patients. Parameter
Placebo
Patients, n
36
38
38
73.3 ⫾ 7.6
76.4 ⫾ 8.1
74.2 ⫾ 9.2
14
20
18
24
18
20
Iris, n Light
32
35
33
Dark
4
3
5
2.2 ⫾ 0.82
2.4 ⫾ 0.86
2.2 ⫾ 0.92
Mean age, y Sex, n Male Female
Mean phaco time Mean phaco %
19.5 ⫾ 5.2
Flurbiprofen Prednisolone
21.6 ⫾ 6.5
20.5 ⫾ 6.6
Means ⫾ SD
are shown in Table 2. Although the mean pupil diameter at baseline tended to be larger in the flurbiprofen and prednisolone groups than in the placebo group, the differences between groups were not statistically significant (P ⫽ .65). The mean pupil size at the subsequent stages of the surgical procedure was invariably smaller than the baseline value in all 3 groups (Table 3 and Figure 2). The paired t test showed that this decrease from baseline was significant at the end of phacoemulsification in all 3 groups. However, at the end of I/A in the prednisolone group (P ⫽ .51) and at the end of surgery in the flurbiprofen group (P ⫽ .22), the reduction was not statistically significant. Intergroup comparison with the 1-way ANOVA showed that overall, there was no statistically significant difference between the 3 groups at any stage of surgery (P ⫽ .21 at the end of emulsification, .37 at the end of I/A, and .16 at the end of the surgery). In the placebo group, the mean decrease in pupil diameter from the baseline ranged from 0.28 to 0.33 mm at various stages of surgery (Table 3). The mean decreases in the other groups were smaller, ranging from 0.08 to 0.17 mm in the flurbiprofen group and 0.16 to 0.22 mm in the prednisolone group. Table 4 shows the 95% CI of the difference between flurbiprofen and prednisolone at each stage of surgery. The latter CI reflects a change in pupil diameter of 0.4 mm, which was considered the smallest clinically significant difference that could affect the conduct of the procedure. The 2 treatment groups were equivalent at all stages except at the end of surgery, when flurbiprofen provided a significantly better mydriasis-maintaining effect than prednisolone. The proportion of patients in each treatment group showing a decrease of 0.4 mm or more from the baseline pupil diameter is shown in Figure 3. The differences between the groups were not statistically significant
Table 2. Mean pupil diameters at various stages of surgery. Mean Diameter (mm) ⫾ SD Group
Before Corneal Incision
End of Emulsification
End of I/A
End of Surgery
Placebo
7.89 ⫾ 0.833
7.56 ⫾ 0.876
7.60 ⫾ 0.844
7.61 ⫾ 0.895
Flurbiprofen
8.08 ⫾ 1.002
7.91 ⫾ 0.997
7.90 ⫾ 1.010
8.00 ⫾ 1.054
Prednisolone
7.92 ⫾ 1.046
7.70 ⫾ 0.991
7.76 ⫾ 1.023
7.73 ⫾ 1.019
I/A ⫽ irrigation/aspiration
2374
J CATARACT REFRACT SURG—VOL 29, DECEMBER 2003
PREDNISOLONE AND FLURBIPROFEN FOR MYDRIASIS DURING PHACOEMULSIFICATION
Table 3. Change in pupil diameter (mm) from baseline. End of Emulsification
End of I/A
End of Surgery
Group
Mean ⫾ SD
P Value
Mean ⫾ SD
P Value
Mean ⫾ SD
P Value
Placebo
⫺0.33 ⫾ 0.366
.0001
⫺0.29 ⫾ 0.397
.0001
⫺0.28 ⫾ 0.462
.0008
Flurbiprofen
⫺0.17 ⫾ 0.301
.0011
⫺0.18 ⫾ 0.337
.0020
⫺0.08 ⫾ 0.397
.027
Prednisolone
⫺0.22 ⫾ 0.483
.0087
⫺0.16 ⫾ 0.483
.051
⫺0.18 ⫾ 0.493
.22
I/A ⫽ irrigation/aspiration
Table 4. The 95% CIs for the difference of means between the fluorbiprofen and prednisolone groups. Stage of Surgery
95% CI
End of emulsification
⫺0.081, ⫹0.170
End of I/A
⫺0.155, ⫹0.108
End of surgery
⫺0.038, ⫹0.0250
CI ⫽ confidence interval; I/A ⫽ irrigation/aspiration
(P ⫽ .32 at the end of emulsification, .29 at the end of I/A, and .33 at the end of surgery). In all groups, miosis of 1.0 mm or more was observed in no more than 5% of patients at any stage. Figure 4 shows the frequency distribution of the different magnitudes of pupil change in the 3 groups at various stages.
Figure 3. (Shaikh) Proportion of patients with miosis 0.4 mm or more from baseline (End Phaco ⫽ end of nuclear phacoemulsification; End A/I ⫽ end of aspiration and irrigation; End surgery ⫽ end of the procedure).
Discussion Pretreatment with various NSAIDs reduces intraoperative miosis during cataract surgery.2 Preoperative treatment is key because once PGs are released, a topical NSAID will not block their effect on the iris. Preoperative treatment with prednisolone acetate also reduces surgical miosis to a limited extent during ECCE.5 In the present study, ANOVA analysis showed that both flurbiprofen and prednisolone failed to have a statistically significant mydriasis-maintaining effect in comparison to the placebo at any stage of surgery. That the patients in all 3 groups in our study had epinephrine in the intraocular irrigating solution could have dictated this outcome. Intraoperative epinephrine has been shown to be the most effective single factor in reducing surgical miosis.6 Our results agree with those of Gimbel, 6 who reports no significant differences among epinephrinetreated eyes with or without pretreatment with NSAIDs.6 Inclusion of treatment groups in which intra-
Figure 4. (Shaikh) Frequency distribution of pupil change from baseline. The y-axis denotes various treatment groups using the following abbreviations: EE is the end of emulsification, EA is the end of aspiration irrigation, and ES is the end of surgery. The x-axis denotes the change in mean pupil diameter using the following abbreviations: ⬎⫽ is more than or equal to, ⬍⫽ is less than or equal to, – is a reduction in diameter, and ⫹ is an increase in diameter.
J CATARACT REFRACT SURG—VOL 29, DECEMBER 2003
2375
PREDNISOLONE AND FLURBIPROFEN FOR MYDRIASIS DURING PHACOEMULSIFICATION
operative epinephrine was not used could have clarified the effects of the drugs used in isolation. However, many phacosurgery units routinely use intraocular adrenaline as a mydriatic agent. This might also explain why the routine use of anti-PGs is not universally accepted. With regard to prednisolone, our results agree with those in a study of ECCE in which prednisolone produced a limited beneficial effect that did not reach statistical significance.5 Our findings reinforce the belief that surgical miosis may be mediated in part by as yet unidentified factors independent of PG pharmacodynamics.7 Also, the differences in dosing regimens, surgical techniques, and patient populations affect the outcomes measure.6 In an intragroup comparison by a paired t test at the crucial stages of cortical cleanup and IOL implantation, both prednisolone and flurbiprofen failed to show a significant antimiotic effect. However, there was an indication at the end of I/A in the prednisolone group of a miosis-inhibiting effect that was statistically significant. This effect could have been spurious as it was not substantiated by the CIs of the difference in the means between the 2 groups. In the flurbiprofen group, significant maintenance of mydriasis at the end of the surgery was suggested by the paired t test and CI data. This observation could be the manifestation of the wellknown phenomenon of synergism between NSAIDs and epinephrine.6 The incidence of miosis of more than 0.4 mm in the flurbiprofen group was about half that in the placebo group. Although the differences between the groups were not statistically significant, this is a clinically significant phenomenon that may be further substantiated by a study repeated on a larger sample. In eyes that are dilated preoperatively with phenylephrine and cyclopentolate and have received no pretreatment with the nonsteroidal drops, Gimbel6 has shown that the mean miosis during phacoemulsification when intraoperative epinephrine was used was 0.2 mm ⫾ 0.059 (SEM). Therefore, the CIs chosen for the equivalence comparison in our study were –0.2 to ⫹0.2 mm. These CIs reflect a change in pupil diameter of 0.4 mm, which we considered the smallest clinically significant difference that could affect the conduct of the procedure. The threshold of pupil inequality clinically detectable to the naked eye is also 0.4 mm. Using the aforementioned equivalence interval, the effect of prednisolone and flurbiprofen was equivalent throughout 2376
the procedure except at the end of surgery, when flurbiprofen gave significantly better effects. Because prednisolone blocks the arachidonic acid metabolism cascade at a higher level, both the cyclooxygenase and lipooxygenase pathways are blocked. The more specific blockage of the cyclooxygenase pathway by the nonsteroidal agent could, in addition to the possible synergism with epinephrine, explain this result. Moreover, the specific mechanism of action of cyclooxygenase blockers such as flurbiprofen may have a variety of biological effects that cannot be satisfactorily explained by inhibition of PG synthesis alone.8 We chose flurbiprofen as a nonsteroidal agent in our study as it has FDA approval for specific use in maintaining mydriasis during cataract surgery. However, newer and more potent NSAIDs such as diclofenac and ketorolac, which are gaining wider acceptance, may have more favorable effects on the pupil and additional benefits with regard to cystoid macular edema (CME).9,10 Prednisolone in combination with ketorolac was shown to have a better effect than either agent alone in the treatment of CME.11 It would be interesting to see whether the similar synergistic antimiotic effect of the combination can be obtained. In our study, uniform and minimal surgical trauma was ensured by excluding cases with a baseline pupil diameter smaller than 6.0 mm. The greater release of inflammatory mediators anticipated in small-pupil surgery may be more susceptible to modification by therapeutic agents. There are also cases in which maintaining mydriasis would benefit the surgeon. Retrobulbar anesthesia was used in our study to achieve uniform anesthesia and akinesia in all patients. Under topical anesthesia, different neuropharmacology considerations may be involved and, therefore, interesting results could be expected in a study using topical anesthesia. An important limitation of our study was that the pretreatment began only 2 hours before surgery. Treatment starting up to 3 days preoperatively is suggested to optimize the antiinflammatory effects of diclofenac.12 Many anterior segment subspecialists start treatment with an NSAID as early as 3 days preoperatively to achieve maximum prophylaxis against intraoperative miosis and postoperative CME. Further study using such a regimen may yield additional and potentially valuable information.
J CATARACT REFRACT SURG—VOL 29, DECEMBER 2003
PREDNISOLONE AND FLURBIPROFEN FOR MYDRIASIS DURING PHACOEMULSIFICATION
References 1. Duffin RM, Camras CB, Gardner SK, Pettit TH. Inhibitors of surgically induced miosis. Ophthalmology 1982; 89:966–997; discussion by SM Poder, 978–979 2. Nichols J, Synder RW. Topical nonsteroidal anti-inflammatory agents in ophthalmology. Curr Opin Ophthalmol 1998; 9(4):40–44 3. Podos SM. Prostaglandins, nonsteroidal anti-inflammatory agents and eye disease. Trans Am Ophthalmol Soc 1976; 74:637–660 4. Kulkarni PS. Steroids in ocular therapy. In: Zimmerman TJ, ed, Textbook of Ocular Pharmacology. Philadelphia, PA, Lippincott-Raven, 1997; 63–67 5. Dube´ P, Boisjoly HM, Bazin R, et al. Comparison of prednisolone acetate and indomethacin for maintaining mydriasis during cataract surgery. Can J Ophthalmol 1990; 25:234–238 6. Gimbel HV. The effect of treatment with topical nonsteroidal anti-inflammatory drugs with and without intraoperative epinephrine on the maintenance of mydriasis during cataract surgery. Ophthalmology 1989; 96:585– 588 7. Flach AJ. Nonsteroidal anti-inflammatory drugs in ophthalmology. Int Ophthalmol Clin 1993; 33(4):1–7 8. To KW, Abelson MB, Neufield A. Nonsteroidal antiinflammatory drugs. In: Albert DM, Jakobiec FA, eds, Principles and Practice of Ophthalmology; Basic Sci-
9.
10.
11.
12.
ences. Philadelphia, PA, WB Saunders, 1994; vol 6, 1022–1027 Diestelhorst M, Schmidl B, Konen W, et al. Efficacy and tolerance of diclofenac sodium 0.1%, flurbiprofen 0.03%, and indomethacin 1.0% in controlling postoperative inflammation. J Cataract Refract Surg 1996; 22: 788–793 Solomon KD, Turkalj JW, Whiteside SB, et al. Topical 0.5% ketorolac vs 0.03% flurbiprofen in inhibition of miosis during cataract surgery. Arch Ophthalmol 1997; 115:1119–1122 Heier JS, Topping TM, Baumann W, et al. Ketorolac versus prednisolone versus combination therapy in the treatment of acute pseudophakic cystoid macular edema. Ophthalmology 2000; 107:2034–2038; discussion by AJ Flach, 2039 Roberts CW. Pretreatment with topical diclofenac sodium to decrease postoperative inflammation. Ophthalmology 1996; 103:636–639
From Christopher Home Eye Unit, Royal Albert Edward Infirmary, Wigan, United Kingdom. Presented at a meeting of the Saudi Ophthalmological Society, Riyadh, Saudi Arabia, March 2002. None of the authors has a financial or proprietary interest in any material or method mentioned.
J CATARACT REFRACT SURG—VOL 29, DECEMBER 2003
2377