Abstract
Purpose
To establish the morphological and functional parameters to predict the effectiveness of intravitreal injections (IVI) of ranibizumab in macular edema due to retinal vein occlusion and to develop a mathematical model for personalized treatment algorithms.
Material and methods
This is a retrospective study of 98 patients (98 eyes) with macular edema, who received IVI of ranibizumab and were followed up for 12 months. Spectral optical coherence tomography scans and best corrected visual acuity (BCVA) assessments were conducted every 3 months. Treatment outcome predictors were calculated based on logistic regression analysis.
Results
The most significant prognostic factors for the long-term BCVA were baseline BCVA (OR 11.1, p = 0.001), foveal volume (OR 10.8, p = 0.001), destruction of external limiting membrane (OR 15.8, p = 0.001), photoreceptor inner/outer segments (OR 11.1, p = 0.001) and retinal pigment epithelium (OR 9.1, p = 0.001). It has also been discovered that post-treatment BCVA correlated with the height of serous retinal detachment (SRD) (r = −0.4, p = 0.001), ganglion cell complex thickness (r = + 0.3, p = 0.01) and focal loss of ganglion cells (r =−0.3, p = 0.005). Patients without SRD required fewer ranibizumab injections (3.8 ± 1.1) for macular edema fluid resorption compared to those with SRD (5.7 ± 1.2, p = 0.03). A mathematical model for predicting and personalized approach therapy of ranibizumab has been obtained (accuracy of 89%).
Conclusion
The effectiveness of IVI of ranibizumab depends on baseline morphological and functional changes. The obtained mathematical model allows for predicting the outcomes of therapy, determining individualized algorithms to increase the treatment effectiveness and to prevent low vision that corresponds to the principles of predictive, preventive, and personalized medicine.
Keywords: Macular edema, Retinal vein occlusion, Hypoxia, Blood-retinal barrier permeability, Endothelial dysfunction, Endothelin-1, Pro-inflammatory, Vasoconstriction, Risk factors, Ranibizumab, Optical coherence tomography, Predictive preventive personalized medicine, Mathematical model, Individualized treatment algorithms, Individual outcomes
Introduction
Retinal vein occlusion (RVO) is a vascular pathology of a visual organ. According to the literature, the main risk factors for RVO include arterial hypertension, dyslipidemia, hyperglycemia, thrombophilia, hyperhomocysteinemia, and increased concentration of VIII factor [1–4].
The pathogenesis of RVO is multifactorial. It has been established that the local vein narrowing in the area of the arteriovenous intersection leads to the violation of hemodynamics, hemorheology with endothelium damage, and its dysfunctioning [5]. Endothelial dysfunction results in the production of angiogenic, thrombogenic, pro-inflammatory, and vasoconstrictor factors. In particular, the overproduction of endothelin-1 leads not only to the mechanical obstruction of the sclerosed vein but also to the increase in retinal venous pressure [6]. In its turn, it can reduce perfusion pressure, which causes microcirculatory disturbances and increases the risk of hypoxia. In addition, an increase in retinal venous pressure leads to an increase in transmural pressure associated with the risk of retinal edema, including in the macular area [6]. The multifactorial effect of mechanical, hemorheological, and immunological mechanisms leads to damage to the blood-retinal barrier and increasing its permeability, followed by development of macular edema [3, 7–12].
One of the key molecules that play a role in the pathogenesis of macular edema due to RVO is vascular endothelial growth factor (VEGF). VEGF is synthesized in vitro and in situ by human retinal pigment epithelium (RPE) cells and is the only endothelial mitogen produced by cultured human RPE in response to hypoxia. VEGF promotes angiogenesis and vascular hyperpermeability by binding to the VEGF receptor 2 which is expressed by the vascular endothelium [2, 13, 14].
There are 3 main reasons to identify VEGF as an important factor in the pathogenesis of macular edema due to RVO: increased expression of VEGF is associated with pathologic transformations of the retinal vasculature, including increased permeability, tissue remodeling and neovascularization; it has been shown that VEGF alone is sufficient to trigger intraocular neovascularization; and, finally, inhibition of VEGF is associated with functional and anatomic improvements in the affected eye. It is these reasons that make this growth factor a “target” for pharmacological intervention [15–18].
Currently, one of the key treatment options for macular edema due to RVO is intravitreal injections of anti-VEGF drugs, such as ranibizumab. Ranibizumab (Lucentis®, Novartis Pharma) is a humanized VEGF antibody fragment that neutralizes all VEGF-A isoforms and their biologically active degradation products. There is a number of prospective, randomized, sham injection-controlled, double-masked, multicenter clinical trials, which show the efficacy and safety of ranibizumab injections in patients with macular edema following RVO and recommend intraocular injections of 0.5 mg ranibizumab as an effective treatment for macular edema with low rates of ocular and non-ocular adverse events [16, 17].
It is recommended that the treatment of macular edema is started with three monthly intravitreal injections of ranibizumab with further dynamic monitoring of clinical and morphological parameters, and, if there is a negative dynamic, additional injections can be administered “as needed” [16–18].
Effectiveness of anti-VEGF therapy depends on a number of factors including the timing of treatment initiation, type of RVO, patient’s age, and presence of concomitant ocular or systemic pathology [19–22]. However, the influence of baseline clinical and anatomical changes on the result of therapy remains insufficiently studied. There is some initial evidence to suggest that baseline best corrected visual acuity (BCVA) plays the most important role in the prognosis of the effectiveness of treatment of macular edema [21, 23], whereas other research suggests the importance of baseline morphometric parameters for the same [19, 24–26].
Prognostic, preventive and individual aspects play an important role among the innovative approaches of modern medicine. The prediction of the therapy efficacy at the onset of the disease is important in the treatment of macular edema with RVO.
Predicting the course of macular edema and the efficacy of its treatment makes it possible to determine the expected therapy duration in order to reduce the economic risks and the possible outcome of the therapy.
A personalized approach is especially relevant in the treatment of macular edema with retinal vein occlusion, when additional injections are prescribed taking into account individual morphological and functional changes that makes it possible to timely adjust the therapy plan and increase patient compliance.
The treatment of macular edema with RVO also requires the prevention of a persistent decline in visual functions, which may lead to the life quality deterioration and possible disablement of patients.
The prediction of the outcome of macular edema therapy and an individual approach help to increase the effectiveness of treatment and to improve targeted prevention of low vision, which corresponds to the basic principles of prognostic, preventive and personalized medicine.
Purpose
The purpose of this study is to determine morphological and functional parameters which can be used to predict the effectiveness of intravitreal injections of ranibizumab in macular edema due to RVO and to develop a mathematical model aimed at determining a personalized treatment algorithm.
Study design
This is a retrospective study conducted using medical records of the ophthalmology department of the regional clinical hospital №3 in Chelyabinsk, Russia. The study was approved by the ethics committee of Federal State Budgetary Educational Institution of Higher Education “South Ural State Medical University” of the Ministry of Health of the Russian Federation.
Material and methods
The retrospective analysis included 98 patients (98 eyes) with macular edema and RVO in the previous 3 months. The average age was 61.6 ± 9.5 years. 61 of the participants were women (62%), and 37 were men (38%). 32 (33%) patients had central retinal vein occlusion (CRVO) and 66 (67%) patients had branch retinal vein occlusion (BRVO).
Inclusion criteria for the study were the following: diagnosis of macular edema due to RVO, baseline BCVA ≤ 20/40 (Snellen equivalent) and central foveal thickness (CFT) ≥ 250 μm at basline. Patients were included in the study only if the baseline data on morphological and functional parameters before treatment (BCVA and OCT) and at least three measurement of BCVA and three OCT scans during the following year were available. OCT scans had to be obtained using the same device (RTVue 100). Only patients treated with at least three intravitreal injections of ranibizumab and followed-up with a Pro Re Nata (PRN) regimen were included.
The exclusion criteria for the study were the following: lack of data on morphological and functional parameters within a year after the start of treatment; inconsistency of the available OCT images or insufficient quality for analysis; the presence of uncompensated glaucoma; ocular surgical interventions in the study eye over the last 3 months, as well as a history of vitreoretinal operations; any eye diseases, that could affect the BCVA and analysis of the results, previous intravitreal injections of aflibercept or dexamethasone.
For baseline, a complete ophthalmologic examination including medical history, BCVA assessed with Snellen visual charts, intraocular pressure (IOP) evaluation by Maklakov’s tonometer, slit lamp biomicroscopy, SD-OCT scans (E MM5, 3D Macular scan, Ganglion Cell Complex (GCC) scan) performed on an RTVue 100/CA (Optovue Inc., USA) was available for all patients included in the analysis.
To minimize the variability of results, only patients evaluated on the same OCT device were included in the analysis.
The intravitreal injections were carried out according to the standard procedure at a dose of 0.5 mg. Re-treatment was performed when a decline in BCVA, an increase of CFT, or an increase or persistence of intraretinal fluid in OCT was observed.
Statistical analysis
Data was analyzed using IBM SPPSS Statistics v 20.0. Kolmogorov-Smirnov tests were employed to check for normality.
Results are expressed as mean and standard deviation (SD) if the variables are continuous and with frequencies and percentages if the variables are categorical.
Baseline variables were compared between treatment groups using Student’s t test and Mann–Whitney U test, Wilcoxon test was used for the continuous cases and with Chi-squared (χ2) and Fisher test for categorical variables. Statistical significance was set at p < 0.05.
The Spearman correlation coefficient was calculated and correlation was considered significant at p < 0.05.
To identify prognostic morphometric parameters that may impact the efficacy of anti-VEGF therapy, the odds ratio and analysis of the 95% confidence interval (CI) was conducted. The following assumptions were used: If the odds ratio (OR) exceeded 1, then it was considered that the study factor has a direct relationship with the probability of the outcome; if the value is less than 1, then the factor was considered to have an inverse relationship with the probability of the outcome; if equal to 1, then the factor has no effect on the probability of the outcome. If both values of the confidence interval lie entirely above or below the null value (OR = 1), it was concluded that the relationship between the factor and the outcome was statistically significant at a significance level of p < 0.05. If the upper and lower limits of the confidence interval embrace the null (risk ration = 1), the results were considered not significant (p > 0.05).
To calculate the probability of obtaining the maximum effect from antiangiogenic therapy, logistic regression analysis was applied, as a result of which a mathematical model was obtained:
where p is the probability of an outcome, e is the mathematical constant (approximately equal to 2.72), and b is the coefficient of the predictor variable X.
Results
Patient population and baseline morphological and functional parameters
To assess the influence of baseline morphometric changes on the efficacy of intravitreal injections of ranibizumab, patients were classified into two macular edema groups based on the findings from the OCT scans. Type 1 was cystoid macular edema (CME), those patients presented multiple cyst-like (cystoid) areas of fluid but no subretinal fluid (Fig. 1) (n = 36). Type 2 comprised of a combination of cystoid macular areas and subretinal fluid with serous retinal detachment (SRD) (Fig. 2) (n = 62).
Fig. 1.
Cystoid macular edema
Fig. 2.
Combination of intraretinal cystoid areas and subretinal fluid
A more detailed description of patient groups and baseline morphological and functional parameters is presented in Table 1.
Table 1.
Characteristics of the study groups and baseline morphological and functional parameters
| Parameters | CME (n = 36) | CME + SRD (n = 62) | p value | |
|---|---|---|---|---|
| Gender | Female (n (%)) | 27 (75%) | 34 (55%) | 0.07 |
| Male (n (%)) | 9 (25%) | 28 (45%) | 0.06 | |
| Average age (years) | 62.9 ± 8.2 | 60.9 ± 10.1 | 0.06 | |
| BCVA (Snellen) | 20/200 ± 20/1000 | 20/200 ± 20/1000 | 0.08 | |
| IOP (mm. Hg) | 20.2 ± 2.1 | 19.9 ± 2.1 | 0.08 | |
| CFT (μm) | 513.4 ± 89.1 | 605.3 ± 193.5 | 0.06 | |
| Fovea volume (FV) (mm3) | 0.3 ± 0.1 | 0.3 ± 0.1 | 0.09 | |
| Average GCC thickness (μm) | 104.3 ± 20.9 | 102.3 ± 20.6 | 0.09 | |
| Superior GCC thickness (μm) | 108.9 ± 26.1 | 101.4 ± 23.9 | 0.07 | |
| Inferior GCC thickness (μm) | 104.0 ± 4.3 | 104.2 ± 22.5 | 0.09 | |
| Focal loss volume (FLV) (%) | 6.4 ± 1.1 | 7.0 ± 0.9 | 0.08 | |
| Global loss volume (GLV) (%) | 9.3 ± 7.4 | 14.5 ± 22.5 | 0.06 | |
| Disruption of external limiting membrane (ELM) integrity (n (%)) | 16 (44%) | 26 (42%) | 0.07 | |
| Disruption of the photoreceptor inner segment/outer segment junction (IS/OS) (n (%)) | 27 (75%) | 50 (81%) | 0.06 | |
| Disruption of retinal pigment epithelium (RPE) (n (%)) | 10 (28%) | 24 (39%) | 0.08 | |
| Choroidal thickness (μm) | 212.6 ± 46.4 | 257.7 ± 6.8 | 0.07 | |
The distribution of age and sex was similar between the groups. Patients in both study groups had poor BCVA, increase of CFT and FV. Average GCC, superior GCC, and inferior GCC thicknesses were within the normative range provided by spectral OCT. GLV was also within the acceptable range, FLV—higher than the normative values (p < 0.05). Destructive changes in the retinal layer were revealed in slightly less than half of the cases in both study groups. There was no statistically significant difference in baseline morphometric parameters between the study groups.
Number of intravitreal injections of ranibizumab
During the 1-year period RVO patients received an average of 5 injections of ranibizumab (range 1–6; mean = 5.8). Patients with CME required significantly fewer injections of ranibizumab for resorption of macular edema (3.8 ± 1.1), than patients with CME + SRD (5.7 ± 1.2) (p = 0.03). Detailed information on the number of intravitreal injections of ranibizumab by group is presented in Table 2.
Table 2.
The number of intravitreal injections of ranibizumab during the 1-year period, n (%)
| Number of injections (n) | CME | CME + SRD | Total |
|---|---|---|---|
| 1 | - | - | - |
| 2 | - | - | - |
| 3 | 36 (37%) | 62 (63%) | 98 (100%) |
| 4 | 23 (23%) | 34 (35%) | 57 (58%) |
| 5 | 9 (9%) | 22 (22%) | 31 (31%) |
| 6 | 4 (4%) | 12 (12%) | 16 (16%) |
All 98 patients received at least three injections of ranibizumab (see Table 2). The maximum number of injections was six (16%); patients with CME + SRD needed significantly more injections than those with just CME (p < 0.05).
Dynamics of morphological and functional parameters after intravitreal injections of ranibizumab
Best corrected visual acuity (BCVA)
Changes in BCVA from baseline are presented in Fig. 3.
Fig. 3.
Change from baseline BCVA (Snellen equivalent). *p < 0.05, significant difference between the study groups (Mann–Whitney U test)
As seen from Fig. 3, after 3 months of ranibizumab treatment, in increase in BCVA was observed in both study groups (p < 0.05). However, patients with CME had significantly higher BCVA after 3 injections compared to patients with CME + SRD (p = 0.003). After 6 months from the start of treatment, there were no significant differences in BCVA between the study groups, but after 12 months of observation in patients with CME had a significantly higher BCVA (p = 0.042).
Central foveal thickness and foveal volume
Mean change in CFT from baseline for each group is provided in Table 3.
Table 3.
Mean change in CFT from baseline
| Time after treatment | Mean (SD) central foveal thickness (CFT), μm | ||
|---|---|---|---|
| CME (n = 36) | CME + SRD (n = 62) | p | |
| Baseline | 513.4 (89.1) | 605.3 (193.5) | 0.001 |
| Month 3 | 318.0 (74.4)* | 312.5 (116.2)* | 0.5 |
| Month 6 | 302.1 (93.7)* | 320.6 (148.3)* | 0.9 |
| Month 12 | 297.8 (104.4)* | 309.2 (161.7)* | 0.7 |
*p < 0.05—in relation to baseline (Wilcoxon test)
After 3 monthly injections of ranibizumab, a two-fold decrease of mean CFT was observed in both study groups, which was statistically significant (p < 0.05). This decrease has remained at the same level in the first year post treatment, and was comparable between groups.
Mean changes in foveal volume from baseline are provided in Table 4.
Table 4.
Mean change in foveal volume from baseline
| Time after treatment | Mean (SD) foveal volume, mm3 | ||
|---|---|---|---|
| CME (n = 36) | CME + SRD (n = 62) | p | |
| Baseline | 0.3 (0.1) | 0.3 (0.1) | 0.7 |
| Month 3 | 0.2 (0.1)* | 0.1 (0.04)* | 0.5 |
| Month 6 | 0.2 (0.1)* | 0.1 (0.04)* | 0.4 |
| Month 12 | 0.1 (0.1)* | 0.1 (0.05)* | 0.7 |
*p < 0.05—in relation to baseline (Wilcoxon test)
After 3 monthly injections of ranibizumab, a statistically significant decrease of mean fovea volume was seen in both study groups (p < 0.05). This decrease has remained almost at the same level in the first year post treatment, with no statistically significant difference between groups.
Influence of baseline morphological and functional parameters on the course of macular edema after intravitreal injections of ranibizumab
We assessed the course of macular edema after intravitreal injections of ranibizumab depending on the baseline morphological changes in the macula. The results are presented in Table 5.
Table 5.
The course of macular edema, n (%)
| Course of macular edema | CME (n = 36) | CME + SRD (n = 62) | p |
|---|---|---|---|
| No recurrence of macular edema | 20 (56%) | 36 (58%) | 0.05 |
| Recurrence of macular edema | 13 (36%) | 14 (23%) | 0.05 |
| Chronic course | 3 (8%) | 12 (19%) | 0.03 |
Patients with CME and patients with CME + SRD presented a comparable rate of recurrence of macular edema after three injections of ranibizumab. However, the duration of macular edema > 6 months without resorption (chronic course) was more often seen in patients with macular SRD.
Predicting the effectiveness of intravitreal injections of ranibizumab
Correlation analysis
At the first stage of the analysis, to determine the relationship between the baseline morphological and functional changes and the effectiveness of intravitreal injections of ranibizumab, we carried out a correlation analysis of the data obtained in the general group of patients with RVO (n = 98). The BCVA during treatment was used to assess the effectiveness of therapy. The data on the relationship between the baseline morphological and functional parameters and BCVA after 3, 6 and 12 months from the start of treatment is presented in Table 6.
Table 6.
Relationship between BCVA during treatment and baseline morphological and functional parameters
| Baseline parameters | BCVA after treatment (Snellen equivalent) | ||
|---|---|---|---|
| Spearman correlation coefficient (r (p)) | |||
| Month 3 | Month 6 | Month 12 | |
| BCVA (Snellen equivalent) | + 0.6 (< 0.001) | + 0.6 (0.001) | + 0.6 (0.001) |
| IOP (mm. Hg) | − 0.2 (0.06) | − 0.4 (< 0.001) | −0.3 (0.002) |
| CFT (μm) | −0.5 (< 0.001) | − 0.4 (0.001) | −0.5 (0.001) |
| Fovea volume (mm3) | −0.5 (< 0.001) | − 0.5 (< 0.001) | −0.5 (0.001) |
| Height of SRD (μm) | −0.3 (0.01) | −0.3 (0.002) | −0.4 (0.001) |
| Average GCC thickness (μm) | + 0.3 (0.018) | + 0.01 (0.3) | + 0.3 (0.01) |
| FLV (%) | −0.3 (0.001) | −0.2 (0.044) | −0.3 (0.005) |
| GLV (%) | + 0.1 (0.2) | + 0.2 (0.03) | + 0.2 (0.058) |
BCVA after Month 3 post treatment significantly correlated with baseline BCVA, CFT, FV, height of SRD, average GCC thickness and FLV (Table 6). A similar trend was observed in the subsequent periods post treatment. In addition, after 6 and 12 months from the start of treatment, it was determined that the BCVA correlated significantly with baseline IOP level.
Based on the results obtained from the correlation analysis, it appears possible to predict the effectiveness of anti-VEGF therapy based on the baseline morphological and functional parameters in routine clinical practice. To support this theory, we performed odds ratio calculations and logistic regression analysis.
The odds ratio of the effectiveness of anti-VEGF therapy for macular edema due to RVO
In order to identify prognostically significant baseline morphological and functional parameters that may affect the effectiveness of anti-VEGF therapy, we calculated the odds ratio (OR) of effectiveness of anti-VEGF therapy. BCVA > 20/40 was taken as an indicator of treatment success. Results are shown in Table 7.
Table 7.
Odds ratio of effectiveness of anti-VEGF therapy
| Baseline parameters | Odds ratio (95% DI for OR) (p) | ||
|---|---|---|---|
| Month 3 | Month 6 | Month 12 | |
| BCVA > 20/200 | 9.1 (2.9–28.8), 0.001 | 7.5 (2.7–20.4), 0.001 | 11.1 (3.5–35.2), 0.001 |
| CFT > 500 μm | 1.1 (0.4–2.7), 0.5 | 0.9 (0.4–2.2), 0.6 | 1.7 (0.7–3.9), 0.2 |
| Fovea volume < 0.3 mm3 | 9.1 (2.5–16.8), 0.001 | 8.8 (2.7–28.1), 0.001 | 10.8 (3.1–39.1), 0.001 |
| SRD | 0.1 (0.7–3.9), 0.3 | 1.1 (0.4–2.5), 0.4 | 1.9 (0.9–4.5), 0.1 |
| Destruction of ELM | 18.7 (5.1–28.3), 0.001 | 20.1 (6.2–64.8), 0.001 | 15.8 (4.9–50.7), 0.001 |
| Destruction of IS/OS | 13.4 (3.7–15.4), 0.001 | 13.6 (4.2–43.5), 0.001 | 11.1 (3.5–35.2), 0.001 |
| Destruction of RPE | 2.2 (1.6–2.9), 0.001 | 10.2 (3.4–34.8), 0.001 | 9.1 (2.9–28.7), 0.001 |
Patients with baseline BCVA > 20/200 and baseline fovea volume < 0.3 mm3 have 9 times higher chance that after three injections of ranibizumab their BCVA will be > 20/40, whereas those with disruption of ELM and IS/OS have a 13–18 times lower chance of > 20/40 BCVA post treatment, and the chance of achieving > 20/40 BCVA post treatment is 2 times lower for those with disruption of RPE (see Table 7). Similar results were obtained for subsequent post treatment periods.
Logistic regression analysis
In order to determine the most significant factors among the baseline morphological and functional parameters that may affect the effectiveness of anti-VEGF therapy for macular edema due to RVO, as well as to obtain a formula for calculating the effectiveness of treatment, a stepwise logistic regression analysis was utilized. This analysis included the initial morphological and functional parameters listed in Table 7. As before, BCVA > 20/40 was used as an indicator of treatment success. The most significant factors that affect BCVA during treatment were identified and are presented in Table 8.
Table 8.
Efficiency predictors of intravitreal injections of ranibizumab
| Clinical parameters | Equation Coefficient (B) | Standard error | p | OR | 95% DI for OR | |
|---|---|---|---|---|---|---|
| BCVA > 20/200 (x1) | 1.325 | 0.28 | 0.001 | 3.8 | 2.1 | 6.5 |
| FV < 0.3 mm3 (x2) | 1.319 | 0.43 | 0.002 | 3.7 | 1.6 | 8.7 |
| Destruction of IS/OS (x3) | − 2.4 | 0.57 | 0.002 | 0.09 | 2.7 | 20.0 |
| Constant value | 1.834 | 0.12 | 0.001 | |||
As follows from Table 8, BCVA after intravitreal injections of ranibizumab is most influenced by the baseline BCVA, foveal volume and disruption of the IS/OS.
To calculate the likelihood of an increase in BCVA to > 20/40 value after three injections of ranibizumab, a logistic regression equation was used:
For X, substitute into the equation: 1—if there is a factor; 2—if there is no factor; x1—baseline BCVA > 20/200; x2—fovea volume < 0.3 mm3; × 3—destruction of IS/OS is presented.
If p < 0.5, then it can be assumed that an increase in BCVA > 20/40 after three injections of ranibizumab will not occur. The sensitivity of the logistic regression analysis was 39.5%, the specificity was 97.8%, and the accuracy was 89%.
Discussion
Taking into account the modern trends in medicine, we focused on the prognostic aspect of determining the therapy effectiveness, and the development of an individual treatment approach and preventive measures aimed at reducing the risk of therapeutic failure.
Predictive diagnostics
The purpose of this study was to determine predictors of the effectiveness of anti-VEGF therapy for macular edema due to RVO. A number of studies have looked at this problem, and the main emphasis was placed on baseline functional and some morphometric parameters [19, 21, 23, 26, 27]. For instance, Jaissle G.B. et al. (2011) noted that one of the most important prognostic factors for the effectiveness of such therapy is the baseline BCVA [21]. Minami Y. et al. (2017) showed that not only the baseline BCVA, but also BCVA 1 day after the injection of ranibizumab is directly correlated with BCVA after 6 months of observation, which can also serve as a prognostic factor for the long-term effect of anti-VEGF therapy [23]. Some authors have established the relationship between the effectiveness of ranibizumab therapy and parameters such as the CRT, FV, the size of the cystic cavities [19, 26, 28]. However, there is no consensus in the literature on this issue, and none of the authors indicate the possibility of involving the inner layers of the retina in the pathological process.
In our study, we found that post-treatment BCVA correlated with the baseline BCVA, CFT, foveal volume, height of SRD, average GCC thickness, and FLV. Based on the results obtained from the correlation analysis we performed odds ratio calculations and logistic regression analysis.
We analyzed the relationship between BCVA after three injections of ranibizumab and in the long-term period (12 months after the start of treatment) on the baseline morphological and functional parameters. It was demonstrated that up to 12 months after treatment the chance of an increase in BCVA > 20/40 was higher with the baseline BCVA > 20/200. This is in contrast to Jaissle G.B. et al. (2011) findings, who noted that the increase in BCVA occurs to a greater extent in patients with low baseline visual acuity [21]. At the same time, other authors emphasized that the higher the baseline BCVA, the better the functional prognosis for the treatment of macular edema with RVO [23]. This discrepancy could be due to methodological variations between the studies, such as patient demographics, follow up period etc.
In addition, among the structural parameters, BCVA after treatment are influenced by: FV < 0.3 mm3, presence of alteration of ELM, presence of alteration IS/OS, disruption of RPE. Those parameters reduce the likelihood of final BCVA > 20/40 both in the early stages (after three injections of ranibizumab), and in the long term (after 12 months). This result can be explained by the fact that the alteration of the retinal layers developing in macular edema can lead to a possible decrease in visual functions [27] with a correlation with the state of vision and OCT parameters at baseline.
A controversial point, according to the literature, is the effect of the presence of SRD in the macula on the effectiveness of anti-VEGF therapy. For instance, Dogan E. et al. (2018) observed that both patients with and without SRD had an increase in BCVA and a decrease of CRT after treatment [25]. Moreover, the best results in terms of retinal structure were observed in patients with SRD, which may be associated with the initially increased CRT in these patients compared with the group of patients without SRD. Other authors also pointed out the increased CRT in patients with SRD [24].
In our study, the BCVA in patients with CME in the long-term period was significantly higher than in patients with CME + SRD. This finding may be associated with the less prominent effect of pro-inflammatory and angiogenic factors in the presence of CME without SRD and, accordingly, the better response to the therapy.
Based on step-by-step logistic regression analysis, we established the main predictors of the effectiveness of therapy are the following: baseline BCVA, FV, and the presence of disruption of IS/OS. Substituting the values of the indicated morphological and functional parameters into the logistic regression formula, the probability of the therapy outcome can be calculated. For instance, with a p value < 0.5, an increase in BCVA > 20/40 after three injections of ranibizumab will not occur with an accuracy of 89% and more injections will be required.
This mathematical model was obtained, which made it possible to predict the effect of anti-VEGF therapy in a particular patient already at the onset of the disease. It allows implementing a comprehensive and systematic approach to the treatment of the disease, considering its economic, social and psychological prospects. This is the predictive role of our results.
Personalisation of medical services
Today, the most important issue to be addressed in macular edema therapy is how to determine the number of injections in each individual case. There is some data in the literature on the dependence of the frequency of recurrences of macular edema on the presence of macular ischemia, peripheral ischemia [29–31], or the levels of cytokines [32, 33]. However, to our knowledge, there is no data on the dependence of the course of macular edema on baseline morphological changes. Meanwhile, it is the baseline morphometric indicators that are readily available at the first visit of the patient and could help to predict the expected number of injections, which would make it possible to personalize the therapy.
In this study, we have demonstrated for the first time that the most important factor determining the course of macular edema after anti-VEGF therapy is the presence of SRD. So, 12 month after treatment, 35% of patients with SRD on PRN regime required 4 additional injections in addition to three initial injections of ranibizumab for complete resorption of macular edema, it was 5 injections for 22% of patients, 6 injections for 12%, in contrast to the group of patients with CME in which 23%, 9% and 4% required 4 additional injections, respectively. Thus, patients without SRD required significantly fewer injections of ranibizumab for complete resorption of macular edema (3.8 ± 1.1) than patients with SRD (5.7 ± 1.2) (p = 0.03). In addition, it was found that the duration of edema for more than 6 months without resorption (chronic course) was more common in patients with SRD. This may be due to the fact that in the presence of not only intraretinal, but also subretinal fluid in the macula, the production of pro-inflammatory and angiogenic factors can be increased to a greater extent, causing relapses or absence of resorption of macular edema, which requires additional injections of the drug.
Thus, the obtained results concerning the effect of morphological parameters on the course of macular edema and the effectiveness of its treatment may be important in choosing a therapy method, predicting the duration and effectiveness of treatment in a particular patient, which makes the approach to treatment more personalized.
Moreover, it is known that the risk factors of stroke, arterial hypertension and retinal vein occlusion are similar [34], and in accordance with the instructions for use of ranibizumab, it should be used with caution in the presence of any risk factors of stroke or previous stroke, due to possible thromboembolic complications. Therefore, an individual approach to specifying a drug regimen is extremely important, and our study was aimed at it.
Targeted prevention
The results of this study and the developed mathematical model allow us to predict the effectiveness of intravitreal administration of ranibizumab, which may significantly increase the treatment effectiveness due to timely adjustment of the treatment plan and the development of an individual approach. The improvement of the treatment effectiveness and the control of macular edema, the chronic course of which may cause irreversible destructive changes in the macular zone, reduce the risk of a persistent decrease in visual functions and low vision. This demonstrates the preventive role of the results of our study.
Limitations
The present study has a number of limitations, the main of which is its retrospective design, as well as the absence of a control group, a small number of patients, and an unbalance between the study groups in terms of the number of patients. Since the aim of our study was to determine morphometric predictors of the efficacy of macular edema therapy, we did not look for patients with fluorescent angiography data and, therefore, could not assess the presence of ischemia and the state of microcirculation, which could affect the efficacy of treatment. In addition, we did not take into account the presence of local factors, in particular the retinal venous pressure, and systemic factors such as the presence of arterial hypertension, dyslipidemia and others, and especially their combination, which could also affect the individual course of the disease. The consideration of these aspects was not the purpose of our study, but these factors can be used in the further study of this problem and for the prognosis of therapy effectiveness.
The strength of the study is the possibility of using the developed mathematical model in clinical practice, when, taking into account the baseline BCVA and OCT findings, already at the first examination of the patient, it becomes possible to predict the outcome of therapy and plan a personalized treatment schedule.
Conclusion
The effectiveness of intravitreal injections of ranibizumab depends on the baseline morphological and functional parameters, including BCVA, CRT, FV, average GCC thickness, FLV, disruption of the ELM integrity, IS/OS and RPE, and presence of SRD. Patients without SRD require 2 times fewer injections of ranibizumab for macular edema resorption than patients with CME + SRD.
To predict the effectiveness of intravitreal injections of ranibizumab, it is possible to use a mathematical model presented in this paper, allowing clinicians to determine an individualized treatment approach, which is consistent with the principles of predictive, preventive, and personalized medicine.
Data availability
Data available on request due to privacy and ethical restrictions.
Code availability
Not applicable.
Declarations
Ethics approval
The study was approved by the ethics committee of Federal State Budgetary Educational Institution of Higher Education "South Ural State Medical University" of the Ministry of Health of the Russian Federation.
Consent to participate
Not applicable.
Consent for publication
Not applicable.
Conflict of interest
The authors declare no competing interests.
References
- 1.Berger AR, et al. Optimal treatment of retinal vein occlusion: Canadian Expert Consensus. Ophthalmologica. 2015;234(1):6–25. doi: 10.1159/000381357. [DOI] [PubMed] [Google Scholar]
- 2.Coscas G, Loewenstein A, Augustin A, Bandello F, Battaglia Parodi M, Lanzetta P, Monés J, de Smet M, Soubrane G, Staurenghi G. Management of retinal vein occlusion-consensus document. Ophthalmologica. 2011;226(1):4–28. doi: 10.1159/000327391. [DOI] [PubMed] [Google Scholar]
- 3.McIntosh RL, Rogers SL, Lim L, Cheung N, Wang JJ, Mitchell P, Kowalski JW, Nguyen HP, Wong TY. Natural history of central retinal vein occlusion: an evidence-based systematic review. Ophthalmology. 2010;117(6):1113–1123.e15. doi: 10.1016/j.ophtha.2010.01.060. [DOI] [PubMed] [Google Scholar]
- 4.Ponto KA, Scharrer I, Binder H, Korb C, Rosner AK, Ehlers TO, Rieser N, Grübel NC, Rossmann H, Wild PS, Feltgen N, Pfeiffer N, Mirshahi A. Hypertension and multiple cardiovascular risk factors increase the risk for retinal vein occlusions: results from the Gutenberg Retinal Vein Occlusion Study. J Hypertens. 2019;37(7):1372–1383. doi: 10.1097/HJH.0000000000002057. [DOI] [PubMed] [Google Scholar]
- 5.Rehak J, Rehak M. Branch retinal vein occlusion: pathogenesis, visual prognosis, and treatment modalities. Curr Eye Res. 2008;33(2):111–131. doi: 10.1080/02713680701851902. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Flammer J, Konieczka K. Retinal venous pressure: the role of endothelin. EPMA J. 2015;6:21. doi: 10.1186/s13167-015-0043-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Daruich A, Matet A, Moulin A, Kowalczuk L, Nicolas M, Sellam A, Rothschild PR, Omri S, Gélizé E, Jonet L, Delaunay K, De Kozak Y, Berdugo M, Zhao M, Crisanti P, Behar-Cohen F. Mechanisms of macular edema: beyond the surface. Prog Retin Eye Res. 2018;63:20–68. doi: 10.1016/j.preteyeres.2017.10.006. [DOI] [PubMed] [Google Scholar]
- 8.Hayreh SS. Retinal vein occlusion. Indian J Ophthalmol. 1994;42(3):109–132. [PubMed] [Google Scholar]
- 9.Hayreh SS, Podhajsky PA, Zimmerman MB. Natural history of visual outcome in central retinal vein occlusion. Ophthalmology. 2011;118(1):119–133.e1-2. doi: 10.1016/j.ophtha.2010.04.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hayreh SS, Zimmerman MB. Branch retinal vein occlusion: natural history of visual outcome. JAMA Ophthalmol. 2014;132(1):13–22. doi: 10.1001/jamaophthalmol.2013.5515. [DOI] [PubMed] [Google Scholar]
- 11.Luna JD, Chan CC, Derevjanik NL, Mahlow J, Chiu C, Peng B, Tobe T, Campochiaro PA, Vinores SA. Blood-retinal barrier (BRB) breakdown in experimental autoimmune uveoretinitis: comparison with vascular endothelial growth factor, tumor necrosis factor alpha, and interleukin-1beta-mediated breakdown. J Neurosci Res. 1997;49(3):268–80. 10.1002/(sici)1097-4547(19970801)49:3%3c268:aid-jnr2%3e3.0.co;2-a. [DOI] [PubMed]
- 12.Muraoka Y, Tsujikawa A, Murakami T, Ogino K, Kumagai K, Miyamoto K, Uji A, Yoshimura N. Morphologic and functional changes in retinal vessels associated with branch retinal vein occlusion. Ophthalmology. 2013;120(1):91–99. doi: 10.1016/j.ophtha.2012.06.054. [DOI] [PubMed] [Google Scholar]
- 13.Campochiaro PA, Bhisitkul RB, Shapiro H, Rubio RG. Vascular endothelial growth factor promotes progressive retinal nonperfusion in patients with retinal vein occlusion. Ophthalmology. 2013;120(4):795–802. doi: 10.1016/j.ophtha.2012.09.032. [DOI] [PubMed] [Google Scholar]
- 14.Pérez-Ruiz M, Ros J, Morales-Ruiz M, Navasa M, Colmenero J, Ruiz-del-Arbol L, Cejudo P, Clária J, Rivera F, Arroyo V, Rodés J, Jiménez W. Vascular endothelial growth factor production in peritoneal macrophages of cirrhotic patients: regulation by cytokines and bacterial lipopolysaccharide. Hepatology. 1999;29(4):1057–1063. doi: 10.1002/hep.510290416. [DOI] [PubMed] [Google Scholar]
- 15.Kamei M, Terasaki H, Yoshimura N, Shiraga F, Ogura Y, Grotzfeld AS, Pilz S, Ishibashi T. Short-term efficacy and safety of ranibizumab for macular oedema secondary to retinal vein occlusion in Japanese patients. Acta Ophthalmol. 2017;95(1):e29–e35. doi: 10.1111/aos.13196. [DOI] [PubMed] [Google Scholar]
- 16.Brown DM, Campochiaro PA, Singh RP, Li Z, Gray S, Saroj N, Rundle AC, Rubio RG, Murahashi WY, CRUISE Investigators Ranibizumab for macular edema following central retinal vein occlusion: six-month primary end point results of a phase III study. Ophthalmology. 2010;117(6):1124–1133.e1. doi: 10.1016/j.ophtha.2010.02.022. [DOI] [PubMed] [Google Scholar]
- 17.Campochiaro PA, Heier JS, Feiner L, Gray S, Saroj N, Rundle AC, Murahashi WY, Rubio RG, BRAVO Investigators Ranibizumab for macular edema following branch retinal vein occlusion: six month primary end point results of phase III study. Ophthalmology. 2010;117(6):1102–1112.e1. doi: 10.1016/j.ophtha.2010.02.021. [DOI] [PubMed] [Google Scholar]
- 18.Larsen M, Waldstein SM, Boscia F, Gerding H, Monés J, Tadayoni R, Priglinger S, Wenzel A, Barnes E, Pilz S, Stubbings W, Pearce I, CRYSTAL Study Group Individualized ranibizumab regimen driven by stabilization criteria for central retinal vein occlusion: twelve-month results. Ophthalmology. 2016;123(5):1101–11. doi: 10.1016/j.ophtha.2016.01.011. [DOI] [PubMed] [Google Scholar]
- 19.Chatziralli I, Theodossiadis G, Chatzirallis A, Parikakis E, Mitropoulos P, Theodossiadis P. Ranibizumab for retinal vein occlusion: predictive factors and long-term outcomes in real-life data. Retina. 2018;38(3):559–568. doi: 10.1097/IAE.0000000000001579. [DOI] [PubMed] [Google Scholar]
- 20.Hirose M, Matsumiya W, Honda S, Nakamura M. Efficacy and visual prognostic factors of intravitreal bevacizumab as needed for macular edema secondary to central retinal vein occlusion. Clin Ophthalmol. 2014;8:2301–2305. doi: 10.2147/OPTH.S74888. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Jaissle GB, Szurman P, Feltgen N, et al. Predictive factors for functional improvement after intravitreal bevacizumab therapy for macular edema due to branch retinal vein occlusion. Graefes Arch Clin Exp Ophthalmol. 2011;249:183–192. doi: 10.1007/s00417-010-1470-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Chung EJ, Hong YT, Lee SC, et al. Prognostic factors for visual outcome after intravitreal bevacizumab for macular edema due to branch retinal vein occlusion. Graefes Arch Clin Exp Ophthalmol. 2008;246:1241–1247. doi: 10.1007/s00417-008-0866-8. [DOI] [PubMed] [Google Scholar]
- 23.Minami Y, Nagaoka T, Ishibazawa Akihiro, Yoshida Akitoshi. Correlation between short- and long-term effects of intravitreal ranibizumab therapy on macular edema after branch retinal vein occlusion: a prospective observational study. BMC Ophthalmol. 2017;17(1):90. doi: 10.1186/s12886-017-0485-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Celık E, Doğan E, Turkoglu EB, Çakır B, Alagoz G. Serous retinal detachment in patients with macular edema secondary to branch retinal vein occlusion. Arq Bras Oftalmol. 2016;79(1):9–11. doi: 10.5935/0004-2749.20160004. [DOI] [PubMed] [Google Scholar]
- 25.Dogan E, Sever O, Köklü Çakır B, Celik E. Effect of intravitreal ranibizumab on serous retinal detachment in branch retinal vein occlusion. Clin Ophthalmol. 2018;17(12):1465–1470. doi: 10.2147/OPTH.S162019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Fujihara-Mino A, Mitamura Y, Inomoto N, Sano H, Akaiwa K, Semba K. Optical coherence tomography parameters predictive of visual outcome after anti-VEGF therapy for retinal vein occlusion. Clin Ophthalmol. 2016;10:1305–1313. doi: 10.2147/OPTH.S110793. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Imai A, Toriyama Y, Iesato Y, Hirano T, Murata T. En-face swept-source optical coherence tomography detecting thinning of inner retinal layers as an indicator of capillary nonperfusion. Eur J Ophthalmol. 2015;25(2):153–158. doi: 10.5301/ejo.5000514. [DOI] [PubMed] [Google Scholar]
- 28.Groneberg T, Trattnig JS, Feucht N, Lohmann CP, Maier M. Morphologic patterns on spectral-domain optical coherence tomography (SD-OCT) as a prognostic indicator in treatment of macular edema due to retinal vein occlusion. Klin Monbl Augenheilkd. 2016;233(9):1056–1062. doi: 10.1055/s-0041-108680. [DOI] [PubMed] [Google Scholar]
- 29.Abri Aghdam K, Reznicek L, Soltan Sanjari M, Klingenstein A, Kernt M, Seidensticker F. Anti-VEGF treatment and peripheral retinal nonperfusion in patients with central retinal vein occlusion. Clin Ophthalmol. 2017;11:331–336. doi: 10.2147/OPTH.S125486. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Choi KE, Yun C, Cha J, Kim SW. OCT angiography features associated with macular edema recurrence after intravitreal bevacizumab treatment in branch retinal vein occlusion. Sci Rep. 2019;9(1):14153. doi: 10.1038/s41598-019-50637-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Jang JH, Kim YC, Shin JP. Correlation between macular edema recurrence and macular capillary network destruction in branch retinal vein occlusion. BMC Ophthalmol. 2020;20(1):341. doi: 10.1186/s12886-020-01611-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Forooghian F, Kertes PJ, Eng KT, Albiani DA, Kirker AW, Merkur AB, Fallah N, Cao S, Cui J, Or C, Matsubara JA. Alterations in intraocular cytokine levels following intravitreal ranibizumab. Can J Ophthalmol. 2016;51(2):87–90. doi: 10.1016/j.jcjo.2015.11.001. [DOI] [PubMed] [Google Scholar]
- 33.Noma H, Yasuda K, Shimura M. Cytokines and recurrence of macular edema after intravitreal ranibizumab in patients with branch retinal vein occlusion. Eur J Ophthalmol. 2019:1120672119885054. 10.1177/1120672119885054. [DOI] [PubMed]
- 34.Polivka J, Jr, Polivka J, Pesta M, Rohan V, Celedova L, Mahajani S, Topolcan O, Golubnitschaja O. Risks associated with the stroke predisposition at young age: facts and hypotheses in light of individualized predictive and preventive approach. EPMA J. 2019;10(1):81–99. doi: 10.1007/s13167-019-00162-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data available on request due to privacy and ethical restrictions.
Not applicable.