Real-World Outcomes of Intraoperative Intravitreal Triamcinolone at Phacoemulsification Cataract Surgery in Eyes with a History of Diabetic Macular Edema Authors: Abigail Sara Affoo, Ronnie Bhola, Vikash Badasee, Vachel Luces, Jeanmarc Sookhoo, Ciara Ramnarine Published: 2026 ABSTRACT ObjectivePostoperative macular edema is a major cause of suboptimal visual acuity in diabetics who have undergone cataract surgery. Intravitreal Triamcinolone Acetonide (IVTA) may be used at the time of cataract surgery with the aim of limiting postoperative macular edema. This study evaluates the real‑world anatomical, visual and safety outcomes of combined intraoperative IVTA and phacoemulsification cataract surgery in eyes with a history of Diabetic Macular Edema (DME).MethodsThis single-centre retrospective observational study analysed the records of 73 eyes, all with a history of DME, which underwent combined phacoemulsification surgery with intraocular lens implantation and IVTA injection between January 2023 and December 2024. Analyses used available paired observations.ResultsThe mean preoperative central subfield macular thickness (CSMT) was 265.0 ± 98.42 µm and 20/73 eyes (27%) had a baseline CSMT of >300 μm preoperatively. OCT scans were available for 51 eyes at 1 month (mean CSMT 313.2 ± 124.04 µm), 52 eyes at 3 months (299.3 ± 112.34 µm), 42 eyes at 6 months (336.2 ± 119.81µm) and 34 eyes at 1 year (310.5 ± 102.11 µm). A significant increase in CSMT was seen at 6 months and 1 year (p<0.01) but not at 1 month and 3 months (p>0.05). 17 eyes received rescue intravitreal therapy within 12 months (10 IVTA, 7 anti-VEGF). Mean visual acuity was 0.988 ± 0.688 logMAR preoperatively with significant improvement at all postoperative visits to 0.414 ± 0.340, 0.436 ± 0.411, 0.464 ± 0.454 and 0.501 ± 0.571 logMAR at 1 month, 3 months, 6 months and 1 year respectively (p<0.001). The mean IOP was not significantly increased at any of the postoperative visits. 3 eyes required the commencement, and 4 eyes required an increased usage of topical ocular anti-hypertensives within 1-year post-surgery. No patient required surgery for uncontrolled postoperative intraocular pressure. There were no cases of endophthalmitis.ConclusionIn this real-world cohort, the findings suggest that intraoperative IVTA may contribute to maintaining CSMT stability during the first three postoperative months in eyes with a history of DME. Introduction Cataract is a leading cause of visual impairment globally.1 Diabetic patients are 2–5 times more likely to develop cataract than non-diabetic patients.2 Cataract surgery which is commonly performed throughout the world has the potential to improve best corrected visual acuity (BCVA) significantly. Patients with diabetic retinopathy have been reported to have poorer postoperative visual acuity and a higher risk of post-operative complications compared to persons without diabetic retinopathy.3 This suboptimal outcome is often due to the occurrence of pseudophakic cystoid macular edema (PCME), the onset or worsening of diabetic macular edema (DME) or the progression of diabetic retinopathy.4 PCME is the most common complication following cataract surgery and has an increased incidence among patients with diabetes, previous retinal vein occlusion and uveitis.5 DME is recognised as a primary cause of vision loss in patients with diabetic retinopathy with it occurring in approximately 1 in every 10 patients with diabetes.3 The severity of DME is associated with increasing severity of diabetic retinopathy, duration of diabetes, poor glycaemic control, diabetic nephropathy, hypertension, dyslipidaemia, pregnancy, uveitis, intraocular surgery and panretinal photocoagulation.6 The incidence of pseudophakic macular edema is estimated at 4% in diabetic patients. Even without retinopathy, persons with diabetes have a relative risk of 1.80 of new macular edema after cataract surgery compared to non-diabetic patients. This relative risk of macular edema postoperatively rises to 6.23 if any diabetic retinopathy is present.7 The presentation of both PCME and DME is with a decline in visual acuity and may include metamorphopsia, loss of contrast sensitivity and central scotomas. On slit lamp biomicroscopy retinal thickening can be seen. Optical coherence tomography (OCT) is useful in the diagnosis and fluorescein angiography may also be used. The incidence of PCME peaks at around 5 weeks postoperatively, and the duration has been reported to range between 72 and 249 days. PCME is often considered a self-resolving condition; however, treatment can accelerate its resolution.8 Treatment options for PCME include topical corticosteroids and non-steroidal anti-inflammatory drugs alone or in combination as a first-line approach with intravitreal corticosteroids and anti-vascular endothelial growth factor agents (anti-VEGF) used when topical therapy is ineffective.9 In contrast, DME once clinically significant frequently requires treatment. The treatment of DME includes the optimisation of glycaemic, hypertensive and lipid control, anti-VEGF therapy, intravitreal corticosteroids, focal/grid laser photocoagulation and surgical intervention.6 The occurrence of post-operative macular edema can have a prolonged negative effect on an individual’s daily functioning and quality of life, and it can also lead to retinal damage with irreversible visual impairment.10 There is also a substantial socio-economic burden from the disease with patients requiring frequent clinic visits, testing such as OCT scans and there is the expense of the treatment medication and administration. For DME, regimes often require multiple injections which is not only costly but is associated with complications from the injections such as endophthalmitis and retinal detachment as well as complications from the medication itself.11 An effective method to prevent macular edema post‑operatively is therefore very desirable. Several methods have been proposed for the prevention of PCME and DME; however, prophylactic treatment practices vary widely between different institutions.5,12 In our setting, patients that are considered at risk of post-operative macular edema following cataract surgery are offered intravitreal triamcinolone acetonide (IVTA) intraoperatively. IVTA is used given that it is less expensive than anti-VEGF agents and steroid implants and that cost remains a large barrier to healthcare in Trinidad and Tobago. Although small prospective trials have evaluated combined IVTA and cataract surgery, real-world data in eyes with a history of DME remain limited particularly in resource constrained settings. This study therefore aimed to describe the anatomical, visual and safety outcomes over one year following combined phacoemulsification and IVTA in eyes with a history of DME. Method This retrospective study analysed the records of patients who underwent combined phacoemulsification cataract surgery with intraocular lens implantation and intravitreal Triamcinolone injection at Trinidad Eye Hospital (TEH), Trinidad and Tobago. To be included in this study, patients were required to be over 18 years old with clinically significant cataract receiving surgery between January 2023 and December 2024 as well as have a history of DME and minimum follow‑up time of 1 month postoperatively. Eyes were considered to have a history of DME once their clinical notes had documentation of the diagnosis made by a doctor in Ophthalmology at any point prior to surgery. Eyes were excluded if there was a history of vitrectomy surgery, planned vitrectomy surgery, a history of uveitis, if there were intraoperative complications, and if they received any Anti-VEGF within 6 weeks prior to cataract surgery or IVTA within 3 months prior to cataract surgery. They were also excluded if they did not have an OCT scan within 3 months prior to surgery or if they did not have any OCT scans postoperatively. All patients received phacoemulsification cataract surgery with intraocular lens implantation into the posterior chamber and Intravitreal Triamcinolone injection at the end of cataract surgery in an operating theatre. Intravitreal Triamcinolone Acetonide suspension 4 mg (40 mg/ml; Tricort 40 manufactured by Cadila Pharmaceuticals Ltd.) was injected 3.5 mm from the superotemporal limbus by use of a 30-gauge needle. All patients received postoperative topical antibiotic/ steroid drops according to local protocol at TEH. Local protocol includes treatment with Ciprofloxacin and Dexamethasone eye drops to the postsurgical eye four times per day starting on the day of surgery with weekly tapering of dosing occurring at around 3–4 weeks postoperatively. Topical NSAIDs are not routinely given to patients at this centre, however, they are prescribed at postoperative visits if PCME develops. Records were analyzed for preoperative risk factors, changes in macular thickness, changes in visual acuity as well as intraoperative and postoperative complications. High risk factors preoperatively included: poor systemic diabetic control (HbA1c>8%), longstanding diabetes, pre-existing DME, proliferative diabetic retinopathy, hard exudates in foveal region on OCT and fellow eye with postoperative macular edema. At this centre, eyes with a history of DME are routinely offered intraoperative IVTA at cataract surgery. SD-OCT (Optovue model: Avanti Scanner and Topcon model Triton) was used to quantify the central subfield macular thickness (CSMT) and parafoveal thickness as well as determine the presence of cysts and other retinal pathology. The CSMT corresponds to the mean macular thickness in the central 1.0 mm area and the parafoveal thickness corresponds to mean thickness of a concentric ring located 1.0 to 3.0 mm around the fovea. Study-defined post-operative macular edema was determined by a CSMT of >300 μm and an increase in CSMT from baseline (pre-operative) of greater than 10%. The baseline (pre-operative) OCT was the patient’s most recent OCT scan prior to surgery. Best-corrected visual acuity (BCVA) was assessed using a logarithm of minimum angle of resolution (logMAR) testing. For the purpose of this study, persons who had a BCVA of counting fingers were assigned a logMAR score of 2.0 for analysis. No eyes were at a visual acuity of hand motion or light perception. Intraocular pressure (IOP) was measured using Tonopen and Icare devices. BCVA and IOP were assessed at each follow‑up visit while OCT scans were done as deemed necessary by the doctor. Follow-up visits were conducted at 1–3 days postoperatively, 1 month postoperatively and then at 3 months, 6 months and 1 year postoperatively as determined by the managing doctor. Data was then analysed using standard statistical software (JASP team 2025 version 0.95.4). Data was reported as count, percent, mean, median and standard deviation. Data demonstrated a non-normal distribution, and therefore non-parametric methods were used accordingly. Paired analysis with Wilcoxon signed rank test required an OCT scan at baseline and at each timepoint. Mean CSMT and parafoveal thickness at each timepoint postoperatively were calculated using the eyes that had an OCT scan at that visit. No imputation was performed for missing OCT data. A p-value of <0.05 was considered as statistically significant. Ethical approval for research was obtained from the ethics committee of the Ministry of Health, Trinidad and Tobago (He: 3/13/441 VoL. II dated December 16th, 2025). Given the retrospective design of the study, the use of anonymised data collected as part of routine clinical care and minimal risk to participants, the requirement of written informed consent was waived. The tenets of the Declaration of Helsinki were adhered to. Results The records of 103 eyes which received phacoemulsification with intraocular lens implantation and intraoperative IVTA were reviewed. Eyes were excluded as per Figure 1 resulting in a total of 73 eyes with a history of DME included in the analysis. 36 (49.3%) eyes were from female patients while 37 (50.7%) were from male patients. The mean age was 68.2 ± 7.33 years. Only 7 eyes (9.6%) had proliferative diabetic retinopathy and 3 eyes (4.1%) had panretinal photocoagulation. 56 eyes (76.7%) had received intravitreal Bevacizumab (IVA) previously of which 18 eyes (24.7%) had also received IVTA previously. Baseline characteristics of the 73 eyes with a history of DME are shown in Table 1. Figure 1Flow diagram showing the number of eyes whose records were reviewed and the number of eyes excluded.Flowchart of eye records reviewed for phacoemulsification, showing exclusions and final analysis count.Table 1Table Showing the Baseline Characteristics of the 73 Eyes with a History of DMECharacteristicValueAge (mean years ± standard deviation)68.2 ± 7.33Male (%)50.7Proliferative diabetic retinopathy present (%)9.6History of panretinal photocoagulation (%)4.1Prior Intravitreal Anti-VEGF treatment (%)76.7Prior Intravitreal treatment with Triamcinolone acetonide (%)24.7Preoperative Central Subfield Macular Thickness (mean µm ± standard deviation)265.0 ± 98.4Best corrected visual acuity (mean logMAR ± standard deviation)0.988 ± 0.688IOP (mean mmHg ± standard deviation)14.62 ± 3.939Abbreviations: DME, diabetic macular edema; IOP, intraocular pressure; Anti-VEGF, anti-vascular endothelial growth factors. Flow diagram showing the number of eyes whose records were reviewed and the number of eyes excluded. Table 1Table Showing the Baseline Characteristics of the 73 Eyes with a History of DMECharacteristicValueAge (mean years ± standard deviation)68.2 ± 7.33Male (%)50.7Proliferative diabetic retinopathy present (%)9.6History of panretinal photocoagulation (%)4.1Prior Intravitreal Anti-VEGF treatment (%)76.7Prior Intravitreal treatment with Triamcinolone acetonide (%)24.7Preoperative Central Subfield Macular Thickness (mean µm ± standard deviation)265.0 ± 98.4Best corrected visual acuity (mean logMAR ± standard deviation)0.988 ± 0.688IOP (mean mmHg ± standard deviation)14.62 ± 3.939Abbreviations: DME, diabetic macular edema; IOP, intraocular pressure; Anti-VEGF, anti-vascular endothelial growth factors. Table Showing the Baseline Characteristics of the 73 Eyes with a History of DME Abbreviations: DME, diabetic macular edema; IOP, intraocular pressure; Anti-VEGF, anti-vascular endothelial growth factors. The mean preoperative CSMT was 265.0 ± 98.4 µm. OCT scans were available for 51 eyes at 1 month (mean CSMT 313.2 ± 124.0 µm), 52 eyes at 3 months (299.3 ± 112.3 µm), 42 eyes at 6 months (336.2 ± 119.8 µm) and 34 eyes at 1 year (310.5 ± 102.1 µm). Paired Wilcoxon signed-rank tests (Holm-adjusted across four timepoints) demonstrated no statistically significant change at 1 month (p=0.386) or 3 months (p=0.964), but significant increases at 6 months (p=0.007) and 1 year (p=0.007). Parafoveal mean thickness showed a similar pattern as seen in Table 2.Table 2Table Showing the CSMT and Parafoveal Thickness at Each TimepointTimeNumber of Eyes with OCT ScanMean CSMT ± SD (µm)Mean Paired Change from Baseline (µm)Median CSMT (µm)p-value Wilcoxon Signed Rank TestHolm-adjusted P ValueMean Parafoveal MT ± SD (µm)Mean Paired Change from Baseline (µm)Median parafoveal MT (µm)p-value Wilcoxon signed rank testHolm-adjusted P valuePre- operative73265.0 ± 98.4—244.0——309.9 ± 59.9—305.0——1 month51313.2 ± 124.042.9269.00.1930.386335.5 ± 80.024.2317.00.1700.3393 months52299.3 ± 112.315.2264.50.9640.964335.4 ± 69.417.6314.00.3280.3396 months42336.2 ± 119.857.4284.50.0020.007360.8 ± 96.342.3334.50.0040.0141 year34310.5 ± 102.167.4269.50.0020.007347.1 ± 82.745.8327.50.0200.061Abbreviations: CSMT, central subfield macular thickness; MT- macular thickness. Table 2Table Showing the CSMT and Parafoveal Thickness at Each TimepointTimeNumber of Eyes with OCT ScanMean CSMT ± SD (µm)Mean Paired Change from Baseline (µm)Median CSMT (µm)p-value Wilcoxon Signed Rank TestHolm-adjusted P ValueMean Parafoveal MT ± SD (µm)Mean Paired Change from Baseline (µm)Median parafoveal MT (µm)p-value Wilcoxon signed rank testHolm-adjusted P valuePre- operative73265.0 ± 98.4—244.0——309.9 ± 59.9—305.0——1 month51313.2 ± 124.042.9269.00.1930.386335.5 ± 80.024.2317.00.1700.3393 months52299.3 ± 112.315.2264.50.9640.964335.4 ± 69.417.6314.00.3280.3396 months42336.2 ± 119.857.4284.50.0020.007360.8 ± 96.342.3334.50.0040.0141 year34310.5 ± 102.167.4269.50.0020.007347.1 ± 82.745.8327.50.0200.061Abbreviations: CSMT, central subfield macular thickness; MT- macular thickness. Table Showing the CSMT and Parafoveal Thickness at Each Timepoint Abbreviations: CSMT, central subfield macular thickness; MT- macular thickness. A sensitivity analysis restricted to the 35 eyes with complete OCT measurements available at baseline, 1 month and 3 months showed no statistically significant change in CSMT from baseline at either 1 month (Holm-adjusted p=0.883) or 3 months (Holm-adjusted p=0.883), consistent with the primary analysis. Preoperatively 20/73 eyes (27%) had a CSMT of >300 μm. Among eyes with available OCT scans at that visit, the number of eyes with a CSMT of >300 μm and an increase in CSMT of >10% from baseline was 10/51 eyes (19.6%) at 1 month, 8/52 eyes (15.4%) at 3 months, 13/42 eyes (31.0%) at 6 months and 9/34 eyes (26.5%) at 1 year. 17 eyes received an intravitreal injection within the year postsurgery of which 10 received IVTA and 7 received an anti-VEGF agent (1.25 mg Bevacizumab). At 1 month postoperatively 1 eye received IVTA, at around 3 months postoperatively 5 eyes received IVTA and 4 eyes received anti-VEGF, at around 6 months postoperatively 2 eyes received IVTA and 1 eye received anti-VEGF and at around 1 year 2 eyes received IVTA and 2 eyes received anti-VEGF. A rescue-censored sensitivity analysis that excluded OCT scans from all visits after an eye received a rescue injection was performed and demonstrated a similar temporal pattern to the primary analysis as shown in Table 3.Table 3Table Showing Mean CSMT at Each Timepoint Excluding Post-Rescue OCT ObservationsTimeNumber of OCT Scans after Rescue CensoringMean CSMT (µm)Mean Paired Change from Baseline (µm)p-value Wilcoxon Signed Rank TestHolm-adjusted P Value1 month51313.2+42.90.1930.3863 months51296.8+11.60.8590.8596 months34327.4+57.20.010.041 year27280.4+54.10.0120.04Abbreviation: CSMT, central subfield macular thickness. Table 3Table Showing Mean CSMT at Each Timepoint Excluding Post-Rescue OCT ObservationsTimeNumber of OCT Scans after Rescue CensoringMean CSMT (µm)Mean Paired Change from Baseline (µm)p-value Wilcoxon Signed Rank TestHolm-adjusted P Value1 month51313.2+42.90.1930.3863 months51296.8+11.60.8590.8596 months34327.4+57.20.010.041 year27280.4+54.10.0120.04Abbreviation: CSMT, central subfield macular thickness. Table Showing Mean CSMT at Each Timepoint Excluding Post-Rescue OCT Observations Abbreviation: CSMT, central subfield macular thickness. Using a definition of post-operative macular edema as a CSMT of >300 μm and an increase in CSMT of >10% from baseline, the presence of several factors were assessed to determine the cumulative risk of developing postoperative macular edema within 1 year. Eyes with foveal exudates had a rate of post-operative macular edema of 48.7% vs 26.5% in those without corresponding to a relative risk (RR) of 1.84 (95% CI: 0.96–3.51). Higher rates of postoperative macular edema were also observed in eyes where the fellow eye had a history of postoperative macular edema, 58.3% in those with vs 34.4% in those without, however, this RR of 1.69 also did not reach statistical significance (95%CI: 0.94–3.06). Very few eyes had proliferative diabetic retinopathy or panretinal photocoagulation inhibiting the determination of reliable relative risk estimates. Duration of diabetes, systemic comorbidities and recent HbA1c levels were not available for the vast majority of patients and, therefore, these risk factors were not able to be analysed. The 73 eyes were subdivided into 2 groups, 20 eyes with a preoperative CSMT of >300 μm and 53 eyes with a CSMT of 300 μm or less. Those with a preoperative CSMT of >300 had a statistically insignificant change in CSMT from baseline at all timepoints postoperatively. Those with a CSMT of 300 μm or less had a statistically insignificant rise in CSMT at 1 month and 3 months but demonstrated a significant rise in CSMT at 6 months and 1 year following Holm correction (see Table 4).Table 4Table Showing the Change in CSMT Postoperatively for Eyes with and without a Preoperative CSMT of >300 μmGroupTimepointNumber of Eyes with OCT ScansMean Change (µm)Median Change (µm)Wilcoxon Signed Rank Test p-valueHolm-adjusted P valuePreop CSMT >300 µm1 month156.1−290.3030.9083 months18−32.7−590.0730.2916 months1430.5−100.7611.0001 year713.1−270.5781.000Preop CSMT ≤300 µm1 month3658.350.0430.0583 months3441.6130.0290.0586 months2870.822<0.001<0.0011 year2781.544<0.001<0.001Abbreviation: CSMT, central subfield macular thickness. Table 4Table Showing the Change in CSMT Postoperatively for Eyes with and without a Preoperative CSMT of >300 μmGroupTimepointNumber of Eyes with OCT ScansMean Change (µm)Median Change (µm)Wilcoxon Signed Rank Test p-valueHolm-adjusted P valuePreop CSMT >300 µm1 month156.1−290.3030.9083 months18−32.7−590.0730.2916 months1430.5−100.7611.0001 year713.1−270.5781.000Preop CSMT ≤300 µm1 month3658.350.0430.0583 months3441.6130.0290.0586 months2870.822<0.001<0.0011 year2781.544<0.001<0.001Abbreviation: CSMT, central subfield macular thickness. Table Showing the Change in CSMT Postoperatively for Eyes with and without a Preoperative CSMT of >300 μm Abbreviation: CSMT, central subfield macular thickness. Mean BCVA preoperatively was 0.988 ± 0.688 logMAR. BCVA changed to 0.414 ± 0.340 logMAR at 1 month, 0.436 ± 0.411 logMAR at 3 months, 0.464 ± 0.454 logMAR at 6 months and 0.501 ± 0.571 logMAR at 1 year, a statistically significant improvement from baseline across all postoperative visits (p<0.001). The mean IOP preoperatively was 14.62 ± 3.939 mmHg. Postoperatively, the mean difference in IOP from preoperative IOP was 0.342 ± 7.159 mmHg at 1–3 days (p= 0.661), −0.648 ± 5.020 mmHg at 1 month (p=0.326), 0.750 ± 5.738 mmHg at 3 months (p=0.491), −2.471 ± 5.005 mmHg at 6 months (p=0.002) and −2.139 ± 4.022 mmHg (p=0.005) at 1 year. 16 eyes were previously treated with an IOP lowering medication for glaucoma, suspect glaucoma or ocular hypertension. 4 of these 16 patients had an additional medication added postoperatively with 3 of these patients experiencing an IOP of greater than 24 mmHg. 3 patients who were never on an IOP lowering medication began treatment, 2 of them at the first postoperative visit and 1 of them at 6 months postoperatively. For all eyes with postoperative IOP elevation, there was subsequent normalisation of IOP with treatment. None required surgery for the elevated IOP. No serious IVTA associated complications such as endophthalmitis were reported. Discussion In the search to optimize postoperative visual outcomes for diabetic patients with cataract, this study reviewed the outcomes of patients with a history of DME receiving combined phacoemulsification and IVTA. Our study noted stabilization of CSMT within the first 3 months followed by a statistically significant increase in CSMT seen at 6 months and 1 year. This effect was particularly noted among those with a preoperative CSMT<300 μm. Phacoemulsification surgery can have an inflammatory effect on eyes. IVTA is thought to prevent or treat macular edema by inhibiting both inflammatory and angiogenic cytokines.3 Intravitreal triamcinolone has a half-life of 18.6 days in non-vitrectomised eyes and IVTA (4 mg) is estimated to have effects on visual acuity and CSMT for 3 months after administration.13,14 All of the eyes in this study had a history of DME with at least 76.7% of eyes having previously required treatment with IVA. Therefore this cohort was considered to be at high risk for the development of postoperative macular edema after cataract surgery. The desired anatomical outcome for these patients postoperatively was for there not to be an increase in macular thickness for those with a preoperative CSMT of <300 μm and for there to be a decrease in macular thickness in eyes with a preoperative CSMT of >300 μm. The lack of a significant increase in CSMT during the first 3 months in this study, corresponding to the estimated duration of IVTA, followed by a significant increase in CSMT at 6 months and 1 year suggests that IVTA may have a positive effect on CSMT stabilisation although temporary. Previous studies supporting the use of IVTA with cataract surgery in diabetic patients include a prospective randomised trial by Addeen et al comparing combined IVTA phacoemulsification surgery to standard phacoemulsification. It showed a significant reduction in CSMT at 3 and 6 months in those receiving IVTA as well as a significantly higher gain in visual acuity in the IVTA group than the control group.4 Another similar study found no significant difference in visual acuity postoperatively between IVTA and control group however the CSMT was significantly lower in treated group than the control at all follow-up visits.15 Both of these randomised trials had a follow up period of 6 months and had less than 22 eyes in the IVTA group. In further support of IVTA, Bassi et al studied the real‑world outcome of 75 eyes with DME receiving combined IVTA cataract surgery and found a reduction in mean CSMT at 3 weeks and 3 months from preoperative visit, however, the decrease was only significant at 3 weeks.16 An additional retrospective study found a significant decrease in CSMT at 1 month postoperatively for both eyes with a history of DME and eyes with active DME.17 These studies contrast to our study in which there was no significant decrease in CSMT at any time point in the year postoperatively even for eyes with a preoperative CSMT of >300 μm. It must be noted, however, that in our study, this >300 μm subgroup was relatively small and results may be skewed by a few very sick eyes that did not respond to IVTA. Still the different postoperative CSMT patterns seen in this study according to baseline macular thickness may suggest that preoperative macular status influences the post-operative anatomical course. Additionally, at this centre, eyes with DME are treated before cataract surgery, which is deferred until the edema has resolved or the CSMT has reached a stable level despite treatment. There was a significant increase in visual acuity from baseline at all timepoints postoperatively. However, much of this improvement would be attributable to the removal of cataract with us unable to quantify the exact effect of IVTA for this study. Our findings are in keeping with previous studies showing the benefits of combined IVTA phacoemulsification surgery yet the poorer prognosis experienced with diabetic eyes compared to eyes without diabetic retinopathy in which an estimated 95% have a visual acuity ≥6/12 (0.3LogMAR).4,15–18 The rates of postoperative macular edema in this study was 19.6% at 1 month, 15.4% at 3 months, 31.0% at 6 months and 26.5% at one year. These cases may represent eyes which are resistant to treatment or nonresponsive to corticosteroid therapy. Small randomised controlled trials have supported the use of anti-VEGF agents in the prevention of postoperative macular edema.19–22 Additionally, the DiMECat study compared the outcomes of patients with DME receiving IVA vs IVTA at the time of cataract surgery in a prospective randomised clinical trial of 61 eyes.23,24 Both IVTA and IVA groups demonstrated improved visual acuity from baseline at 1 year postoperatively. At 6 months the IVTA group had sustained a reduction in CSMT compared to the IVA group although this effect was not maintained at 1 year. The need for additional treatments postoperatively was higher in the IVA group with 43% at 6 months and 53% at 1 year not requiring treatment compared to 76% at 6 months and 62% at 1 year not requiring treatment for the IVTA group. Another major factor in the choice of IVTA rather than IVA for the prevention of macular edema is that IVTA is considerably less expensive. This study is to the authors best knowledge, the first study investigating the prevention of postoperative macular edema after cataract surgery by the addition of IVTA in the Caribbean population. Cost is a major barrier to healthcare in Trinidad and Tobago with anti-VEGF agents unavailable at several public hospitals. There is a lack of data on the combined use of anti-VEGF agent and IVTA for the prevention of macular edema post cataract surgery, however, recent evidence has not shown a benefit in combined IVTA and anti-VEGF vs anti-VEGF monotherapy in the treatment of DME.25 It may still be considered for select eyes given the rates of postoperative macular edema found in our study, but further research would be required to test its efficacy and identify which eyes might benefit. Concerns with intravitreal IVTA include the risk of increased IOP and endophthalmitis. Mean IOP was not significantly elevated at any of the post-operative visits. Furthermore, of the 57 eyes that were not previously on IOP lowering medication only 3 eyes (5.3%) required the commencement of treatment. Other studies show similar cases of eyes requiring the commencement of topical therapy but overall, IVTA is considered safe.4,15–17,26 Nevertheless, cautious use remains advised in those eyes with glaucoma or ocular hypertension. Although we had no cases of endophthalmitis, it remains a possible severe complication. By administering IVTA in an operating theatre with control of the eye, we believe the risks associated with IVTA injection to be less than when conducted in a clinic setting and therefore in those eyes with a high likelihood of developing macular edema requiring injection, they should receive it at the time of surgery. A strength of this study is that it provides real‑world data on the outcomes of eyes receiving combined IVTA and cataract surgery in the Caribbean population which has a notably high prevalence of diabetes. Also the long follow up period of 1 year provides good data on safety. The main limitation of this study is the lack of a control group who did not receive IVTA with cataract surgery which prevents a full understanding of how postoperative visual acuity is affected from the IVTA compared to the effect of cataract surgery. The study’s retrospective design relied on medical records in which there was incomplete documentation, for example the authors of this study had access to the records of patients for their treatment at this centre but did not have access to records of treatment patients may have received at a different centre previously. Another limitation of this study’s retrospective design is the lack of OCT scans at all post-operative visits. Nevertheless, we believe that our study group was large enough to have a sufficient number of OCT scans at each visit to allow a satisfactory comparison of CSMT and parafoveal thickness at all timepoints. Conversely, the lack of OCT scans at each visit could represent a selection bias as affected eyes or on the other hand those that are adherent and doing well may be more likely to present for OCT scanning and therefore the patients with available imaging may not be representative of the entire cohort. A further limitation of the study is that 2 OCT devices were used since TEH has 2 clinic locations with one device at each location. The differences in measurement between devices are likely to be small and more importantly, patients used the same machine for all their OCT scans allowing true comparison in CSMT at different timepoints for the same eye. Additionally, some eyes received IVTA or anti-VEGF treatment during the one year postoperatively. This treatment represents a source of confounding particularly after 3 months postoperatively as these additional treatments would have acted to decrease CSMT and parafoveal thickness within the 3 months after administration. Therefore, there is likely an underestimation of the number of true cases of macular edema in this study occurring after the 3 months postoperatively. Given these limitations, there is a need for a large randomised controlled prospective study without missing data. Conclusion Postoperative macular edema among diabetic patients remains a tremendous cause of poor visual acuity after cataract surgery. In this real-world study of eyes with a history of DME, there was stabilisation of CSMT within the first 3 months postoperatively while significant increases were observed at later follow-up. As no control group was included, these findings describe the postoperative course of this cohort rather than establishing a treatment effect, however, the early stability coinciding with the expected duration of the IVTA administered suggests that intraoperative IVTA may be associated with short-term CSMT stability. Meaningful visual improvement and an acceptable safety profile were observed. In resource-limited settings, intraoperative IVTA may represent a potential low-cost strategy to tackle the burden of postoperative macular edema in eyes with a history of DME receiving phacoemulsification surgery, however a large randomised controlled prospective study is required to confirm these findings. Patients with a history of DME should receive close follow-up after cataract surgery since a substantial proportion may require rescue treatment in the first year postoperatively.