The PREMED study took place at 12 European surgical centres and involved 914 non-diabetic and 213 diabetic patients.
All patients in the study received standard phacoemulsification for cataract and placement of an IOL with intraoperative and postoperative antibiotics administered according to local protocols.
Patients all required phacoemulsification cataract surgery in at least one eye. They were excluded if the had functional monoculus; previous surgery of the study eye; increased risk for developing CME; contra-indication for study medication; macular pathology that may influence visual acuity; posttraumatic cataract; or the use of steroids, NSAIDS or anti-glaucoma medication.
In the non-diabetic population, the 914 patients received either a topical NSAID (bromfenac 0.09%) or a topical corticosteroid (dexamethasone 0.1%), or a combination of both.
Central subfield macular thickness (CSMT) was 9.6 µm higher in the dexamethasone group at 6 weeks than in the combination treatment group, which was statistically significant (P = 0.002).
The CSMT in the dexamethasone group at 6 weeks was 6.9 µm higher than in the bromfenac group, but this result fell just short of the threshold for statistical significance. (A P value of 0.026 versus a Bonferroni correction threshold of 0.025). None of differences in CSMT among the groups was statistically significant at 12 weeks.
The pattern in parafoveal thickness was similar at 6 and 12 weeks.
Total macular volume and perifoveal thickness followed the same pattern at 6 weeks, but the thickness in the dexamethasone group remained higher at 12 weeks by these parameters.
Compared with the combination of the two drugs, the risk of developing CME (defined as increase in CSMT of at least 10% as compared with baseline and cystic changes on optical coherence tomography) was 2.6 times higher for the bromfenac group, and 3.7 times higher for the dexamethasone group.
The incidence of CSME (CME with less than 0.2 logMAR CDVA improvement as compared with baseline) was 1.5% in the combination group compared with 3.6% for bromfenac alone and 5.1% for dexamethasone alone.
The researchers treated 213 patients with diabetes differently. They excluded those with severe diabetic retinopathy or diabetic macular oedema, but included those with mild or moderate disease.
All of the patients with diabetes received the combination of bromfenac and dexamethasone, but one group received no additional treatment; one group received a subconjunctival injection with 40 mg triamcinolone acetonide; one group received an intravitreal injection with 1.25 mg bevacizumab; and one group received or a combination of both after cataract surgery.
The macular thickness and volume in the diabetic patients was significantly lower in patients who received a subconjunctival injection with triamcinolone compared with patients who did not. Patients who received a subconjunctival injection of triamcinolone had a 12.3µm lower central subfield thickness at 6 weeks, which was highly significant (P = 0.007). And the difference was 9.7 µm at 12 weeks, which was also highly significant (P = 0.014).
No patient who received subconjunctival triamcinolone developed CME, while intravitreal bevacizumab had no significant effect on macular thickness.
However, patients receiving triamcinolone also experienced a higher risk of increased intraocular pressure (IOP). A 6 weeks, the patients receiving triamcinolone had, on average, a 1.7 mmHG higher IOP than the patients not receiving the steroid. At 12 weeks they had a 2.5 mmHG higher IOP (p < 0.001).
In his own practice, Dr Nuijts said he would generally be willing to take the small risk of raising a patient’s IOP in return for the high probability of preventing CME.