In this study, the OCT parameters CRT and TMV as markers for the reduction of macular edema, RNFLT as a parameter for the integrity of the nerve fibers, visual acuity and eye pressure with different numbers of anti-VEGF injections from 120 patients with neovascular AMD in the clinical setting were examined.
CRT and TMV are commonly used for objective documentation of progress, whereas the BCVA represents the main outcome parameter of most studies.
Macular edema is significantly reduced after injection series up to 18 injections and in a constant size range of approximately 88–64 µm (CRT) or 0.75–0.55 mm3 (TMV). A slight improvement in the visual acuity of 0.06 logMAR can be observed initially. With further necessary injections, only stabilization is achieved compared to the initial visual acuity. This implies only a temporary improvement in visual acuity despite response to anti-VEGF therapy, corresponding to randomized controlled studies ANCHOR, IVAN, MARINA and CATT [6, 7, 18]. The last mentioned study could even figure out even a loss of BCVA in long-term follow-ups .
Similar to these studies, no differences between different anti-VEGF agents could be found .
However, the comparability of this study with the randomized controlled trials mentioned above is limited due to the lack of sharply defined treatment periods in this study.
As illustrated in the meta-analysis by Mehta et al. , numerous studies were published since the establishment of anti-VEGF injections as a clinical standard for nAMD showing less strong effects compared with the registration studies. The results of the German study population of the AURA study group have shown that in clinical real-world setting only a temporary, slight improvement in visual acuity is achieved after the loading phase . Almost all examinations (> 98.7%) in the clinical setting have been irregular with longer intervals than 5 weeks. The importance of regular controls (4-weekly intervals) and the indication of regular OCT examinations has been pointed out by Ziemssen et al. More frequent sequences of controls and injections have a positive effect on improving and maintaining visual acuity during anti-VEGF therapy, which was demonstrated by a comparison of national differences in health systems with different effectiveness in therapy .
As the most commonly used objective parameter, CRT was included as a criterion for progression in studies determining the intervals of therapy [3,4,5].
Furthermore, individual risk factors for treatment response were figured out . High age and high CRT at baseline predicted high CRT reduction, whereas more injections, treatment with ranibizumab, and male sex predicted a low CRT reduction .
The magnitudes of CRT reduction among all studies differ a lot: In the SUSTAIN study population, an average CRT reduction of 101 µm after the loading phase was observed (baseline = 340.5 µm ± 113 µm), which remained constant in the further course of therapy . In the PrONTO study, the mean CRT reduction after 12 months was 216 µm with a baseline of 394 µm . In the EXCITE study, CRT was quantified 96–106 µm depending on the therapy regimen with baseline values of 314–325 µm (± 85–119 µm) . A subsumption of the results is limited due to the strong fluctuations in the CRT measurement between the listed studies and this one. Unlike Gabai et al. who pointed out a correlation between BCVA and CRT after 12 months in a real-world setting, the present study achieves a sufficient reduction in edema, whereas BCVA improvement was observed only after 3 injections . Although the influence of CRT on the baseline BCVA is known, its significance during the course of therapy is discussed controversial .
A retrospective clinical study from Germany achieved similar results to the present study . This study underlines that visual impairment and a purely quantitative determination of CRT (− 100 µm) are not sufficient as a criterion of progression, but that morphological criteria should always be taken into account .
Total macular volume (TMV) is defined as the sum of all nine areas of the ETDRS grid. The reduction in TMV is significantly between − 0.55 and 0.75 mm3 for all anti-VEGF agents combined. TMV is more unconventional than CRT as a parameter for assessing therapeutic success, so that only a few studies are available. Ma et al. analyzed in a clinical setting 118 patients under the PRN regimen with bevacizumab . The baseline TMV in this study is 10.2 ± 2.0 mm3. Unlike the present study, a strong reduction of 2.07 mm3 was observed after 9 month. The lower baseline value in our study (8.73–9.09 mm2) and the lower treatment adherence of our cohort due to the real-world study design can be discussed as possible reasons.
It serves as an additional marker for macular edema besides CRT. An advantage of TMV is the detection of CNV within a larger retinal area. However, this results in the disadvantage of a smaller discrimination range in which alterations take place. The differences between these studies underline that the TMV should be interpreted carefully as a progression parameter.
IOP and RNFL
Increases in intraocular pressure during anti-VEGF therapy have been described in different studies . A meta-analysis from 2015 determines 3.5–11.6% as the average IOP increase with a defined clinically relevant magnitude of > 21 mmHg and a minimum increase of 20% from the baseline value . In our study, no significant changes in IOP could be observed. This meta-analysis included patients with glaucoma, whereas in our cohort these patients were excluded. A retrospective study showed that patients with pre-existing glaucoma experienced higher rates of elevated IOP when compared with patients without pre-existing glaucoma (33% vs. 3.1%, respectively; p < 0.001) . The same study described a higher incidence of sustained elevation of IOP after bevacizumab. An explanation for this fact and elevated IOP levels in glaucoma patients after anti-VEGF injections in general could be that high molecular weight proteins, such as bevacizumab (150 kDa), which is approximately three times larger than ranibizumab may obstruct in the outflow and accumulate in the trabecular meshwork .
The small numbers of cases in the bevacizumab groups post-15 and post-18 with 6 and 5 eyes showing a significantly reduced baseline RNFLT limits the validity. The reduction of RNFLT (− 1.4 µm) in post-3 ranibizumab patients (n = 52) can be emphasized and discussed more in detail.
In meta-analysis of 2016 including 288 eyes, treated with different agents in a Pro-Re-Nata regime, no significant changes of RNFL were found in total . In a subgroup analysis carried out in the same study, dividing the 288 eyes in subgroups with different numbers of injections [≥ 10 injections (n = 70) and < 10 injections (n = 218)], a significant decrease (MD = − 0.250, 95% CI − 0.441 to − 0.058, p = 0.011) was pointed out, similar to our observations in the post-3 group of ranibizumab. However, no long-term effects on RNFL with increasing numbers of injections of ranibizumab were neither observed in this study nor in several prospective designed studies using ranibizumab (n = 20–30) [33, 34]. Because of the elevated susceptibility for RNFL damage in glaucoma patients, these patients were excluded in our study. In regard to our findings, further prospective, well-designed, longitudinal examinations are needed to investigate the exact behavior of RNFL changes after ranibizumab injections.
As a retrospective study, the present study is limited in many respects. The number of patients decreases in groups of higher injection numbers, which is related to the low availability of patient data with long-term anti-VEGF therapy. As a result, systematic bias occurs at higher injection numbers: Patients in whom no therapeutic success can be achieved are underrepresented in these groups, because no therapy adherence is guaranteed under these circumstances. Furthermore, multimorbid and elderly patients, who represented the majority in our cohort, are less likely to be on long-term therapy. Similarly, patients in groups with higher injection numbers are mostly those with better baseline factors and favorable treatment outcome. The small number of patients in high injection numbers generally limits the significance of these results. Furthermore, it has to be pointed out that the number of injections correlates only slightly with the duration of therapy. However, this is urgently needed for an assessment of this chronic disease. The influence of the irregular therapy intervals is thus indirectly included in the overall outcome of the study. Statements about therapy intervals and therapy regimens were not considered in this study.