Comparison of Optical Coherence Tomography Biomarkers between Bevacizumab Good Responders and Nonresponders Who were Switched to Dexamethasone Implant in Diabetic Macular Edema

Article information

Korean J Ophthalmol. 2023;37(2):137-146

doi : https://doi.org/10.3341/kjo.2022.0109

Jeong Hyun Lee, Joo Young Shin, Jeeyun Ahn

Department of Ophthalmology, Seoul Metropolitan Government Seoul National University Boramae Medical Center, Seoul National University College of Medicine, Seoul, Korea

Corresponding Author: Jeeyun Ahn, MD, PhD. Department of Ophthalmology, Seoul Metropolitan Government Seoul National University Boramae Medical Center, Seoul National University College of Medicine, 20 Boramae-ro 5-gil, Dongjak-gu, Seoul 07061, Korea. Tel: 82-2-870-2416, Fax: 82-2-831-0714, E-mail: autre24@gmail.com

Received 2022 August 10; Revised 2023 January 18; Accepted 2023 February 13.

Abstract

Purpose

To compare volumetric optical coherence tomography (OCT) biomarkers in bevacizumab responsive and bevacizumab refractory diabetic macular edema (DME) patients switched to the dexamethasone implant to ultimately identify possible prognostic indicators.

Methods

Retrospective analysis of DME patients treated with bevacizumab were done. Patients were divided into those who showed response to bevacizumab (bevacizumab only group) and others who were switched to the dexamethasone implant due to lack of response to bevacizumab (switching group). Volumetric OCT biomarkers such as central macular thickness (CMT), inner and outer cystoid macular edema (CME) volume, serous retinal detachment (SRD) volume, retinal volume (CME + SRD volume) within the 6-mm Early Treatment of Diabetic Retinopathy Study circle were calculated. OCT biomarkers were followed up throughout treatment.

Results

Among total of 144 eyes, 113 patients were included in the bevacizumab only group and 31 patients were included in the switching group. Compared to the bevacizumab only group, the switching group showed higher baseline CMT (558.00 ± 209.60 μm vs. 454.96 ± 125.88 μm, p = 0.003), larger inner CME (6.02 ± 1.43 mm³ vs. 5.12 ± 0.87 mm³, p = 0.004) and SRD volume (0.32 ± 0.40 mm³ vs. 0.11 ± 0.09 mm³, p = 0.015) and higher proportion of patients with SRD (58.06% vs. 31.86%, p = 0.008). In the switching group, CMT, inner CME and SRD volume all showed significant reduction after switching to the dexamethasone implant.

Conclusions

DME with large SRD and inner nuclear layer edema volume may be more effectively treated with the dexamethasone implant than bevacizumab.

Keywords: Bevacizumab; Dexamethasone implant; Diabetic macular edema; Optical coherence tomography biomarker

Diabetic retinopathy (DR) is the leading cause of visual impairment in the working age population [1]. Among many clinical features of DR, diabetic macular edema (DME) is the most common cause of vision deterioration due to its effect on central vision [2]. DME pathophysiology involves the release of vascular endothelial growth factor (VEGF) and other inflammatory mediators, such as interleukin 1β (IL-1β), IL-6, IL-10, tumor necrosis factoralpha (TNF-α), monocyte chemotactic protein-1 (MCP-1), and interferon-induced protein-10 (IP-10), causing the breakdown of vascular barriers and accumulation of fluid [3]. Although anti-VEGF agents are currently the first-line treatment of choice for DME, intravitreal dexamethasone implant (0.7 mg; Ozurdex, Allergan Inc) has also demonstrated effectiveness in treating persistent DME [4]. In clinical practice, intravitreal dexamethasone implant injection may be considered a more ideal choice in pseudophakic eyes, those refractory to anti-VEGF drugs or patients hesitant to receiving frequent intravitreal injections.

Based on optical coherence tomography (OCT), DME may be classified into three different morphologic subtypes; diffuse retinal thickening (DRT), cystoid macular edema (CME), and serous retinal detachment (SRD) [5,6]. SRD is believed to be cause by the breakdown of the outer blood-retinal barrier (BRB), in association with retinal inflammation, whereas DRT and CME are thought to be mainly caused by breakdown of the inner BRB [7–10]. Therefore, DME may be further classified into two different pathogenic subtypes; SRD and non-SRD (DRT and CME) [11]. Due to the anti-inflammatory property of corticosteroids, some studies have advocated the use of dexamethasone intravitreal implants over anti-VEGF agents in treating SRD type of DME [12–15].

In the current study, clinical manifestations and volumetric OCT biomarkers of DME patients who were responsive to bevacizumab, and others switched to the dexamethasone implant due to lack of response, were analyzed. Dexamethasone implant was the initial choice over aflibercept mainly due to the cost and recommended 5 monthly loading dose which may be burdensome to the patient in a real-world clinic setting. OCT biomarkers were analyzed to determine features indicative of responsiveness to either anti-VEGF or steroid therapy in DME. Changes in inner and outer CME were analyzed separately as the inner and outer retina have distinct blood supplies, and the inner retina in diabetic patients show distinctive features compared with the outer retina. Inner retinal thinning has been described as an early sign of neurodegeneration in patients with diabetes and inner retinal swelling was found to corelate better with best-corrected visual acuity (BCVA) [16–18].

Materials and Methods

Ethical statements

This was a retrospective study performed with approval from the Institutional Review Board of Seoul Metropolitan Government Seoul National University (SMG-SNU) Boramae Medical Center (No. 20-2021-104), and in accordance with the Declaration of Helsinki. The requirement for informed consent was waived due to the retrospective nature of the study.

Participants

Medical records of treatment-naïve DME patients who received intravitreal bevacizumab injection at SMG-SNU Boramae Medical Center between September 2015 and April 2021 were reviewed. If both eyes of a patient were treated, the eye initially treated was selected. Patients were divided into two groups based on their responsiveness to bevacizumab, evaluated by the change in central macular thickness (CMT), and measured with OCT: patients who responded to bevacizumab (bevacizumab only group) and patients who were switched to dexamethasone implant due to lack of response to bevacizumab (switching group). Bevacizumab only group was further divided in to two subgroups: patients successfully treated with single bevacizumab injection (group 1), and those who required more than one bevacizumab injection (group 2). The subgroups were compared with the switching group (group 3) to determine whether groups 2 and 3 share common features that differentiate them from group 1. The exclusion criteria were as follows: macular edema due to causes other than DR (e.g., age-related macular degeneration, vascular occlusion, postsurgical macular edema); significant media opacity affecting OCT quality (e.g., dense cataract, corneal opacity, vitreous hemorrhage); epiretinal membrane evident on OCT; history of uveitis; history of intraocular surgery (within 6 months prior to the initial injection); YAG laser capsulotomy (within 2 months prior to the injection); and panretinal photocoagulation (within 3 months prior to the initial injection).

Treatment protocol

When DME identified on OCT and CMT, computed at the center of Early Treatment of Diabetic Retinopathy Study (ETDRS) 1-mm circle on OCT volume scan, was greater than 300 μm, the patient was considered for bevacizumab injection. After 1 month, if CMT reduction was more than 50 μm or 10% of initial thickness, additional bevacizumab injection was considered, taking BCVA, CMT, hemoglobin A1c (HbA1c), and patient compliance into account. If CMT increased or CMT reduction was less than 50 μm or 10% of initial thickness, switching to the dexamethasone implant was considered. Patients who received only bevacizumab was labeled as “bevacizumab only group” and patients who were switched to dexamethasone implant was labeled as “switching group.”

Dexamethasone implant was injected 1 month after the last bevacizumab injection. Patients were followed up at 1 and 3 months postinjection to examine intraocular pressure (IOP) and CMT, and the necessity of an additional injection was determined after 3 months. Baseline was defined as the day of first bevacizumab intravitreal injection and the endpoint was defined as the follow-up day for the last intravitreal injection. Intravitreal injections were performed in a clean room as an outpatient procedure mostly on the day of the patients’ visit to the hospital for OCT examination. After disinfection and sterile draping, intravitreal injection of bevacizumab (1.25 mg/0.05 mL) using a 30-gauge needle or dexamethasone implant (0.7 mg implant of dexamethasone; Ozurdex, Allergan Inc) was done under topical anesthesia with 0.5% proparacaine hydrochloride (Alcaine, Alcon).

Data collection

Clinical features such as the number of injections, age, sex, BCVA, IOP, duration of diabetes mellitus, HbA1c, presence of chronic kidney disease, lens status, DR severity, and follow-up period were collected. Follow-up period was defined as the period between the patients’ initial injection and the day of last follow-up. Adverse events after intravitreal dexamethasone injection such as elevated IOP, cataract formation, endophthalmitis, or retinal detachment were also noted.

Measurement of OCT biomarkers

OCT examinations were performed using spectral domain OCT (Spectralis, Heidelberg Engineering). Macular volume scans consisting of 31 horizontal high-speed B-scans covering 30° × 25° centered on the foveola were acquired at each visit. Retinal volume was measured from autosegmented values within the Spectralis viewer as the volume between the internal limiting membrane and Bruch’s membrane (BrM). Autosegmentation lines delineating each individual retina layer were then adjusted to calculate CME and SRD volume. SRD volume was calculated as the volume between BrM and inner border of SRD. CME volume was calculated by subtracting SRD volume from retinal volume. Inner and outer CME were divided by the border between the inner nuclear layer (INL) and outer plexiform layer. A representative case is presented in Fig. 1A–1E.

Fig. 1

The case of a 57-year-old male patient in the switching group who was treated for diabetic macular edema in his right eye. (A,B) Infrared images with retinal thickness color overlay. (C–E) horizontal optical coherence tomography scans. (C,D) The autosegmentation lines delineating the external limiting membrane (ELM) and retinal pigment epithelium (RPE) have been adjusted in the right eye to obtain the volumes of the serous retinal detachment and the inner and outer cystoid macular edema (CME) components (blue, red, and green color overlays). (E) Note the original segmentation lines in his left eye. ILM = internal limiting membrane; BrM = Bruch’s membrane; SRD = serous retinal detachment.

Statistical analysis

Numerical data are expressed as mean and standard deviation, and categorical variables as absolute frequency and percentage. Normal distribution of examined variables was verified by the Kolmogorov-Smirnov test. For each parameter, we performed statistical comparisons between groups using the Student t-test or Mann-Whitney U-test for numerical variables and Pearson chi-squared test or linear-by-linear association for categorical variables. Pretreatment and posttreatment OCT biomarker values were compared with the Wilcoxon signed-rank test. Statistical analyses were performed using IBM SPSS ver. 27.0 (IBM Corp). A p-value of <0.05 was considered statistically significant.

Results

A total of 144 eyes of 144 patients were included in this study, of whom 113 patients were responsive to bevacizumab (bevacizumab only group) and 31 patients were switched to the dexamethasone implant due to lack of response (switching group). Bevacizumab only and switching groups received average 2.04 ± 1.40 and 3.03 ± 2.50 injections of bevacizumab, respectively, and the switching group received additional 1.29 ± 0.69 injections of the dexamethasone implant. In the switching group, 25 patients (80.6%) received single injection of dexamethasone implant and six patients (19.4%) received more than one injection (range, two to four injections). Baseline demographics of the bevacizumab only and switching groups were compared and there were no significant differences (Table 1).

Table 1

Baseline demographics (n = 144)

When baseline OCT biomarkers between the two groups were compared, the switching group showed significantly larger CMT, inner CME and SRD volume, and higher proportion of patients with SRD than the bevacizumab only group (Table 2). CMT of patients without SRD was not significantly different between the two groups.

Table 2

Comparison of baseline optical coherence tomography biomarkers

The bevacizumab only group was divided into two subgroups according to the number of bevacizumab injections (single vs. more than two injections). A total of 54 and 59 patients were included in groups 1 and 2, respectively. When baseline OCT biomarkers of groups 1, 2, and 3 were compared, there were significant differences in CMT, inner CME and SRD volume and proportion of patients with SRD (Table 3). Subgroup analyses showed inner CME volume of groups 1 and 2 to be significantly smaller than that of group 3. SRD volume of group 1 was significantly smaller than that of group 2. Proportion of SRD patients increased as patients received additional bevacizumab and as patients were switched to the dexamethasone implant.

Table 3

Comparison of baseline optical coherence tomography biomarkers when divided into three groups

Changes in OCT biomarkers and BCVA before and after treatment were investigated and are summarized in Table 4. For the bevacizumab only group, parameters prior to the first bevacizumab injection and 1 month after the last injection, and for the switching group, those prior to the first and 1 month after the last bevacizumab injection and 3 months after Ozurdex injection, were compared.

Table 4

Optical coherence tomography biomarkers and BCVA measured throughout treatment in both groups

BCVA of the switching group did not significantly change after bevacizumab injection (p = 0.416) but improved significantly after switching to the dexamethasone implant (p = 0.001). All OCT biomarkers in the bevacizumab only group showed significant reduction 1 month after the last bevacizumab injection. In the switching group, all OCT biomarkers significantly decreased 3 months after switching to the dexamethasone implant. OCT biomarkers that failed to show response to bevacizumab injection in the switching group were CMT, inner CME and SRD volume in patients with SRD. Retinal, CME, and outer CME volume all showed significant reduction after 1 month after bevacizumab injection. Representative photographs of OCT images of each group throughout treatment are shown in Fig. 2A–2J.

Fig. 2

Images of (A–D) bevacizumab only group and (E–J) switching group. (B,D,F,H,J) Representative optical coherence tomography images. (A,B,E,F) At baseline. (C,D,G,H) After bevacizumab injection. (I,J) After dexamethasone implant injection. (E–J) Persistent macular edema after bevacizumab injection regressed after switching to dexamethasone implant.

Of the 31 patients who were switched to the dexamethasone implant, adverse events were reported in five patients (16.1%). IOL elevation of more than 25 mmHg at least once during the follow-up period was shown in five patients (16.1%) and were all managed successfully with topical medications. Increase in cataract density by 2 grades or more from baseline was reported in two patients (6.5%). Other more severe adverse event such as endophthalmitis or retinal detachment were not reported. No adverse events were found in the bevacizumab only group.

Discussion

With the introduction of anti-VEGF treatment and OCT, monitoring patient response to injections has been focused on changes in CMT. In this article, we evaluated volumetric OCT biomarkers outside the central subfield but within ETDRS 6-mm grid, to cover the whole macula area. Switching group showed higher CMT, SRD volume, and higher rate of SRD type of DME at baseline compared to the bevacizumab only group. Throughout the course of DME treatment, SRD volume decreased after switching to the dexamethasone implant in the switching group. Since CMT is measured at the central 1 mm of ETDRS and SRD is present mainly at the central fovea, higher CMT may have been affected by the higher rate of SRD type of DME and larger SRD volume.

There have been previous studies explaining the differential effect of various intravitreal agents on SRD and non-SRD type of DME [10,19–22]. In agreement with our study, these studies advocate anti-VEGF agent as being more effective in non-SRD type of DME and corticosteroids more effective in SRD type of DME. Of several previous studies, Demircan et al. [22] manually measured SRD height to quantitatively analyze OCT biomarkers and Bayat and Elcioglu [13] included non-SRD patients as a control group. Other studies lacked quantitative analysis of SRD and a control group. Our study includes volumetric measurement of OCT biomarkers and bevacizumab only group as control group to complement previous studies advocating higher efficacy of dexamethasone implant in SRD type of DME .

In addition to SRD volume, inner CME volume was larger in the bevacizumab refractory group at baseline and decreased after switching to the dexamethasone implant. The inner and outer plexiform layers have been reported to present physical resistance to fluid movements within the retina [23]. Thus, inner and outer CME volume defined in this study could represent edematous changes of the INL and outer nuclear layer. The intermediate and deep capillary plexuses are located above and below the INL [24,25]. Elnahry et al. [26] compared the deep capillary plexus vessel density between bevacizumab responsive and refractory groups and found significantly greater vascular density in the bevacizumab responsive group. As such, low capillary plexus density may have led to the lack of response to bevacizumab but response to corticosteroids. Further studies regarding correlation between vessel density and responsiveness to corticosteroid using OCT angiography could be done.

Retinal volume at baseline did not show any significant difference between the two groups unlike CMT. As CMT is more likely to be affected by SRD and center-involving lesions, it may be more informative to analyze changes in retinal volume. In the switching group, CMT significantly decreased after switching to the dexamethasone implant whereas retina volume significantly decreased after bevacizumab injection. This is due to the fact that SRD involved a limited portion of the area within the ETDRS 6-mm circle in SRD type of DME, and OCT biomarkers outside the SRD involving area responded to bevacizumab leading to reduction of retinal volume. Hodzic-Hadzibegovic et al. [27] measured retinal volume outside the central subfield in DME patients and 57% of CMT nonimprovers showed retinal volume reduction outside the central subfield, which also supports our findings.

Since BCVA is mainly determined by the central 1-mm subfield, BCVA change followed a similar pattern as with CMT and SRD. VA gain between the two groups were not significantly different. VA gain in SRD type of DME is controversial. Lower VA due to SRD type of DME at baseline may be the reason for more favorable visual acuity gain according to previous studies [28,29]. Other studies in which baseline BCVA was comparable between SRD and non-SRD type of DME showed similar improvements in BCVA which was also the case in our study [12,13].

As patients receive more bevacizumab injections and eventually switch to the dexamethasone implant, they may be considered as receiving more intensive treatment. However, there were no significance differences in CMT and retinal volume when group 1 versus group 2 and group 2 versus group 3 were compared. Rate of patients with SRD increased as patients received more intensive treatment. This may imply that qualitive analysis such as the presence or absence of SRD may be as important as quantitative OCT parameters like CMT and retinal volume. SRD volume of group 1 was even smaller than group 2 which may imply that bevacizumab responsive patients with larger SRD volume could require multiple number of injections. When treating patients with large volume of SRD, dexamethasone implant may be considered as the initial treatment option.

In this study, patients were switched earlier to dexamethasone implant compared to other studies. Other studies performed consecutive anti-VEGF injections (range, three to eight injections) and then evaluated the responsiveness [12,13,30,31]. Consecutive injection without evaluation of treatment effect could lead to unnecessary injection and increased financial burden for the patient. By switching early to the dexamethasone implant, specific OCT biomarker responsive to each treatment modality could have been more elaborately identified. Dexamethasone implant was reinjected at an interval of 3 months in our study which is shorter than 4 months of BEVORDEX study [32]. Since no optimal regimen for dexamethasone implant exists and patients were refractory to bevacizumab in our study, 3 months seemed to be reasonable. Rate of patients with elevated IOP (16.1%) and cataract progression (6.5%) in our study was relatively lower than that of BEVORDEX study (46% for elevated IOP and 43% for cataract progression). This is probably due to fewer injection number and shorter follow-up period.

Baseline demographics showed no difference regarding rate of proliferative diabetic retinopathy (PDR). Hobbs et al. [33] suggested that DME and PDR are different disease entity driven by separate molecular mediators which could lead to altered responsiveness to anti-VEGF agents when treating DME. However, the rate of PDR was similar for both groups in this study.

This study has several limitations. First, the retrospective design resulted in variable bevacizumab injection treatment schedules before switching to the dexamethasone implant. Secondly, specific retreatment criteria were not established. Finally, short-term follow-up period after injection may have led to the underestimation of adverse events and limited evaluation of the long-term treatment effect. However, we were able to gather a moderate number of patients with a strict injection schedule, add a control group, and analyze volumetric OCT biomarkers in DME subcompartments, enabling better understanding of prognostic OCT biomarkers in DME.

In conclusion, DME with large SRD volume and INL edema volume at baseline may be more effectively treated with dexamethasone implant rather than bevacizumab.

Acknowledgements

None.

Notes

Conflicts of Interest: None.

Funding: None.

References

1. Williams R, Airey M, Baxter H, et al. Epidemiology of diabetic retinopathy and macular oedema: a systematic review. Eye (Lond) 2004;18:963–83.

2. Moss SE, Klein R, Klein BE. The 14-year incidence of visual loss in a diabetic population. Ophthalmology 1998;105:998–1003.

3. Klaassen I, Van Noorden CJ, Schlingemann RO. Molecular basis of the inner blood-retinal barrier and its breakdown in diabetic macular edema and other pathological conditions. Prog Retin Eye Res 2013;34:19–48.

4. Haller JA, Kuppermann BD, Blumenkranz MS, et al. Randomized controlled trial of an intravitreous dexamethasone drug delivery system in patients with diabetic macular edema. Arch Ophthalmol 2010;128:289–96.

5. Otani T, Kishi S, Maruyama Y. Patterns of diabetic macular edema with optical coherence tomography. Am J Ophthalmol 1999;127:688–93.

6. Kim BY, Smith SD, Kaiser PK. Optical coherence tomographic patterns of diabetic macular edema. Am J Ophthalmol 2006;142:405–12.

7. Sonoda S, Sakamoto T, Yamashita T, et al. Retinal morphologic changes and concentrations of cytokines in eyes with diabetic macular edema. Retina 2014;34:741–8.

8. Vujosevic S, Torresin T, Berton M, et al. Diabetic macular edema with and without subfoveal neuroretinal detachment: two different morphologic and functional entities. Am J Ophthalmol 2017;181:149–55.

9. Daruich A, Matet A, Moulin A, et al. Mechanisms of macular edema: beyond the surface. Prog Retin Eye Res 2018;63:20–68.

10. Seo KH, Yu SY, Kim M, Kwak HW. Visual and morphologic outcomes of intravitreal ranibizumab for diabetic macular edema based on optical coherence tomography patterns. Retina 2016;36:588–95.

11. Hu Y, Wu Q, Liu B, et al. Comparison of clinical outcomes of different components of diabetic macular edema on optical coherence tomography. Graefes Arch Clin Exp Ophthalmol 2019;257:2613–21.

12. Castro-Navarro V, Cervera-Taulet E, Navarro-Palop C, et al. Intravitreal dexamethasone implant Ozurdex in naïve and refractory patients with different subtypes of diabetic macular edema. BMC Ophthalmol 2019;19:15.

13. Bayat AH, Elcioglu MN. Effects of dexamethasone treatment on serous retinal detachment in ranibizumab-resistant diabetic macular edema. Ther Adv Ophthalmol 2020;12:251584142097. 1936;

14. Ceravolo I, Oliverio GW, Alibrandi A, et al. The application of structural retinal biomarkers to evaluate the effect of intravitreal ranibizumab and dexamethasone intravitreal implant on treatment of diabetic macular edema. Diagnostics (Basel) 2020. 10p. 413.

15. Bonfiglio V, Reibaldi M, Pizzo A, et al. Dexamethasone for unresponsive diabetic macular oedema: optical coherence tomography biomarkers. Acta Ophthalmol 2019;97:e540–4.

16. van Dijk HW, Verbraak FD, Kok PH, et al. Early neurodegeneration in the retina of type 2 diabetic patients. Invest Ophthalmol Vis Sci 2012;53:2715–9.

17. Chhablani J, Sharma A, Goud A, et al. Neurodegeneration in type 2 diabetes: evidence from spectral-domain optical coherence tomography. Invest Ophthalmol Vis Sci 2015;56:6333–8.

18. Ebneter A, Wolf S, Abhishek J, Zinkernagel MS. Retinal layer response to ranibizumab during treatment of diabetic macular edema: thinner is not always better. Retina 2016;36:1314–23.

19. Kim M, Lee P, Kim Y, et al. Effect of intravitreal bevacizumab based on optical coherence tomography patterns of diabetic macular edema. Ophthalmologica 2011;226:138–44.

20. Shimura M, Yasuda K, Yasuda M, Nakazawa T. Visual outcome after intravitreal bevacizumab depends on the optical coherence tomographic patterns of patients with diffuse diabetic macular edema. Retina 2013;33:740–7.

21. Ozdemir H, Karacorlu M, Karacorlu SA. Regression of serous macular detachment after intravitreal triamcinolone acetonide in patients with diabetic macular edema. Am J Ophthalmol 2005;140:251–5.

22. Demircan A, Ozkaya A, Alkin Z, et al. Comparison of the effect of ranibizumab and dexamethasone implant on serous retinal detachment in diabetic macular edema. J Fr Ophtalmol 2018;41:733–8.

23. Antcliff RJ, Hussain AA, Marshall J. Hydraulic conductivity of fixed retinal tissue after sequential excimer laser ablation: barriers limiting fluid distribution and implications for cystoid macular edema. Arch Ophthalmol 2001;119:539–44.

24. Provis JM. Development of the primate retinal vasculature. Prog Retin Eye Res 2001;20:799–821.

25. Snodderly DM, Weinhaus RS, Choi JC. Neural-vascular relationships in central retina of macaque monkeys (Macaca fascicularis). J Neurosci 1992;12:1169–93.

26. Elnahry AG, Noureldine AM, Abdel-Kader AA, et al. Optical coherence tomography angiography biomarkers predict anatomical response to bevacizumab in diabetic macular edema. Diabetes Metab Syndr Obes 2022;15:395–405.

27. Hodzic-Hadzibegovic D, Sander BA, Valerius M, Lund-Andersen H. Diabetic macular edema treated with anti-vascular endothelial growth factor: considerations related to nonimprovers. Ophthalmol Retina 2018;2:1133–42.

28. Dugel PU, Hillenkamp J, Sivaprasad S, et al. Baseline visual acuity strongly predicts visual acuity gain in patients with diabetic macular edema following anti-vascular endothelial growth factor treatment across trials. Clin Ophthalmol 2016;10:1103–10.

29. Qu-Knafo LM, Fajnkuchen F, Sarda V, et al. Impact of serous retinal detachment on the efficacy of ranibizumab in diabetic macular oedema. Acta Ophthalmol 2017;95:e434–5.

30. Al-Latayfeh M, Abdel Rahman M, Shatnawi R. Outcome of single dexamethasone implant injection in the treatment of persistent diabetic macular edema after anti-VEGF treatment: real-life data from a tertiary hospital in Jordan. Clin Ophthalmol 2021;15:1285–91.

31. Hsia NY, Lin CJ, Chen HS, et al. Short-term outcomes of refractory diabetic macular edema switch from ranibizumab to dexamethasone implant and the influential factors: a retrospective real world experience. Front Med (Lausanne) 2021;8:649979.

32. Gillies MC, Lim LL, Campain A, et al. A randomized clinical trial of intravitreal bevacizumab versus intravitreal dexamethasone for diabetic macular edema: the BEVORDEX study. Ophthalmology 2014;121:2473–81.

33. Hobbs S, Mehta A, Patel A, Das A. Proliferative diabetic retinopathy and diabetic macular edema: are these two different disease processes? A retrospective cross-sectional study. Invest Ophthalmol Vis Sci 2016;57:1592.

Article information Continued

This is an Open Access journal distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Demographic	Bevacizumab only group (n = 113)	Switching group (n = 31)	p-value
No. of injections			0.500*
Bevacizumab	2.04 ± 1.40	3.03 ± 2.50
Dexamethasone implant	-	1.29 ± 0.69
Sex			0.092^†
Male	65 (57.5)	23 (74.2)
Female	48 (42.5)	8 (25.8)
Age (yr)	62.45 ± 11.38	65.06 ± 9.45	0.243^‡
BCVA (logMAR)	0.55 ± 0.42	0.50 ± 0.37	0.466*
Intraocular pressure (mmHg)	14.21 ± 3.55	14.71 ± 3.61	0.585*
Diabetes mellitus duration (yr)	15.60 ± 8.36	16.92 ± 10.04	0.643*
Hemoglobin A1c (%)	7.87 ± 1.46	7.58 ± 1.24	0.383*
No. of patients with CKD	20 (17.7)	9 (29.0)	0.163^†
Lens status			0.446^†
Phakic	74 (65.5)	18 (58.1)
Pseudophakic	39 (34.5)	13 (41.9)
Diabetic retinopathy			0.325^†
PDR	55 (48.7)	12 (38.7)
Non-PDR	58 (51.3)	19 (61.3)
Follow-up period (mon)	19.65 ± 5.96	21.61 ± 5.60	0.120*

Variable	Bevacizumab only group (n = 113)	Switching group (n = 31)	p-value
CMT (μm)	454.96 ± 125.88	558.00 ± 209.60	0.003*
CMT in non-SRD type patients (μm)	441.51 ± 107.67	489.92 ± 256.30	0.840*
Retina volume (mm³)	11.11 ± 1.67	12.56 ± 3.17	0.065*
CME volume (mm³)	11.08 ± 1.65	12.36 ± 3.06	0.078*
Inner CME (mm³)	5.12 ± 0.87	6.02 ± 1.43	0.004*
Outer CME (mm³)	6.02 ± 1.03	6.37 ± 1.97	0.866*
No. of patients with SRD	36 (31.9)	18 (58.1)	0.008^†
CME volume (mm³)	11.61 ± 1.69	12.78 ± 2.10	0.091*
SRD volume (mm³)	0.11 ± 0.09	0.32 ± 0.40	0.015*

Variable	Bevacizumab only group		Switching group	Group comparisons (p-value)

	Group 1 (n = 54)	Group 2 (n = 59)	Group 3 (n = 31)	1 vs. 2 vs. 3	1 vs. 2	2 vs. 3	1 vs. 3
No. of bevacizumab injections	1.00 ± 0	3.00 ± 1.35	3.03 ± 2.46	-	-	-	-
Central macular thickness (μm)	423.44 ± 82.20	483.80 ± 150.52	558.00 ± 209.60	0.003*	0.081^†	0.062^†	0.004^†
Retina volume (mm³)	10.87 ± 1.38	11.33 ± 1.88	12.56 ± 3.17	0.169*	-	-	-
CME volume (mm³)	10.83 ± 1.37	11.29 ± 1.85	12.36 ± 3.06	0.198*	-	-	-
Inner CME (mm³)	5.07 ± 0.70	5.16 ± 0.99	6.02 ± 1.43	0.016*	0.688^†	0.011^†	0.008^†
Outer CME (mm³)	5.89 ± 0.80	6.13 ± 1.20	6.37 ± 1.97	0.909*	-	-	-
No. of patients with SRD	18 (33.3)	20 (33.9)	18 (58.1)	0.043^‡	-	-	-
CME volume (mm³)	11.13 ± 1.19	12.10 ± 1.99	12.78 ± 2.10	0.157*	-	-	-
SRD volume (mm³)	0.07 ± 0.04	0.14 ± 0.10	0.32 ± 0.40	0.020*	0.013^†	0.161^†	0.005^†

Variable	Bevacizumab only group			Switching group

	Preinjection	1 mon after last bevacizumab injection	p-value*	Preinjection	1 mon after last bevacizumab injection	3 mon after Ozurdex injection	p-value*

							Preinjection vs. 1 mon after last bevacizumab injection	1 mon after last bevacizumab injection vs. 3 mon after Ozurdex injection
BCVA (logMAR)	0.54 ± 0.41	0.42 ± 0.37	<0.001	0.50 ± 0.37	0.46 ± 0.27	0.34 ± 0.21	0.416	0.001
CMT (μm)	452.29 ± 123.08	331.38 ± 62.84	<0.001	558.00 ± 209.60	540.06 ± 186.54	342.19 ± 83.68	0.433	<0.001
Retina volume (mm³)	11.11 ± 1.68	10.07 ± 1.06	<0.001	12.56 ± 3.17	12.09 ± 3.06	9.97 ± 1.51	0.017	<0.001
CME volume (mm³)	11.07 ± 1.66	10.06 ± 1.06	<0.001	12.36 ± 3.06	11.94 ± 2.93	9.94 ± 1.50	0.013	<0.001
Inner CME (mm³)	5.11 ± 0.86	4.68 ± 0.66	<0.001	6.02 ± 1.43	5.92 ± 1.44	5.19 ± 1.00	0.150	<0.001
Outer CME (mm³)	6.03 ± 1.04	5.39 ± 0.67	<0.001	6.37 ± 1.97	6.07 ± 1.79	4.76 ± 0.83	0.002	<0.001
Patients with SRD
CME volume (mm³)	11.66 ± 1.69	10.44 ± 0.92	<0.001	12.78 ± 2.10	12.15 ± 2.08	10.16 ± 1.21	0.022	<0.001
SRD volume (mm³)	0.11 ± 0.09	0.03 ± 0.10	<0.001	0.32 ± 0.40	0.23 ± 0.40	0.04 ± 0.13	0.102	0.019