Article Text
PDF
Abstract
Purpose To compare the effect on contrast sensitivity (CS) measurements of panretinal photocoagulation (PRP) associated with intravitreal bevacizumab (IVB) injections versus PRP alone in high risk proliferative diabetic retinopathy (HR-PDR).
Design Prospective, randomised, masked, controlled trial.
Participants 42 patients with HR-PDR with visual acuity ≥20/200.
Methods Eyes were randomised to one of two groups: one underwent PRP and IVB injections (study group) and the other PRP alone (control group). PRP was performed three times during the study and IVB injection was administered twice.
Main outcome measures Mean change in CS threshold scores between and within groups, from baseline to 6 months.
Results Seven patients presented with vitreous haemorrhage and were removed from the study. Mean results for CS threshold (at 1.5, 3, 6, 12 and 18 cycles per degree (cpd) frequencies) for patients with and without diabetic macular oedema showed no significant differences (p>0.05 for all comparisons) between the two groups. In 35 eyes in the control group, compared with baseline values, there was significant worsening (p<0.05) of CS at 1.5, 12 and 18 cpd after 1 month, at 12 cpd after 3 months, and at 6 and 12 cpd after 6 months. In the study group, there was significant improvement in CS at 3 cpd, 3 months after treatment.
Conclusions In eyes with HR-PDR, PRP treatment is associated with deterioration of CS while adjuvant use of bevacizumab prevents such deterioration. CS evaluation seems to support the adjuvant use of bevacizumab when using PRP for the treatment of HR-PDR.
ClinicalTrials.gov Identifier NCT 01389505.
- Angiogenesis
- Retina
- Treatment Lasers
- Visual perception
- Treatment Medical
Statistics from Altmetric.com
Introduction
Consensus exists that panretinal photocoagulation (PRP) is the treatment of choice for patients with severe diabetic retinopathy (DR), including those with high risk proliferative DR (HR-PDR).1 ,2 The Diabetic Retinopathy Study demonstrated a reduction of 50% or more in severe visual loss for patients with HR-PDR who received PRP.3
However, previous studies have shown that PRP may be complicated by the development or worsening of macular oedema (MO),4 with transient or persistent reduction of visual acuity (VA).
While VA is the most commonly used psychophysical test, it has many drawbacks as it evaluates only optotypes with high degrees of contrast while in the real world objects have different degrees of variability in contrast and spatial frequency.5 Contrast sensitivity (CS) testing allows measurement of the patient’s ability to see low contrast patterns and provides more information on visual function than VA.6 As a result, including CS testing in clinical trial protocols may provide a more complete picture of the effect of treatment on visual function.
DR characteristically involves all retinal layers and consequently affects both VA and CS, and dissociation of VA and CS in the initial stages has been reported previously.7 Little is known, however, about the effects on CS of more advanced stages of DR, such as HR-PDR.
Bevacizumab (Avastin, Genentech Inc, San Francisco, California, USA) is a full length humanised monoclonal antibody that inhibits all isoforms of the vascular endothelial growth factor A family.8 Intravitreal bevacizumab (IVB) injection leads to satisfactory visual results when used in patients with diabetic MO (DMO),9 and recent studies have evaluated the effect on VA of IVB injection associated with PRP in the treatment of PDR.10–12 In cases of PDR without DMO, some studies did not confirm a significant difference in VA measurements between groups.10 ,11 On the other hand, when PDR is associated with DMO, adjuvant use of IVB injection to PRP can contribute to improved VA.12
While several studies have compared the results of PRP treatment alone with PRP associated with IVB injection on VA, none has estimated the effects of treatment on CS. We therefore designed a comparative clinical trial to investigate the effect on CS of treating HR-PDR eyes with and without DMO with either PRP alone or PRP combined with IVB injections.
Methods
The trial was an interventional, prospective, masked, randomised study, conducted between February 2011 and June 2012. The study followed the principles of the Declaration of Helsinki and was approved by the institutional review board of the ethics committee. All participants gave informed consent.
Forty-two patients (14 women) with a diagnosis of HR-PDR with or without DMO were enrolled in the study. All patients underwent a complete endocrine evaluation and ophthalmic examination, including VA, CS and intraocular pressure measurements, slit lamp examination and fundoscopy.
CS measurements were performed under photopic conditions (85 cd/m2) with the VCTS 6500 chart (Vistech Consultants Incorporation, Dayton, Ohio, USA) positioned 3 m away from the patients eyes, allowing evaluation of five spatial frequencies with sine wave grating charts: 1.5, 3.0, 6.0, 12.0 and 18.0 cycles per degree (cpd).
Patients also underwent an optical coherence tomography (OCT) evaluation using Stratus OCT equipment (Carl Zeiss Ophthalmic Systems Inc; Humphrey Division, Dublin, California, USA), and DMO was defined when OCT foveal thickness (FT) was ≥250 μm.13
Digital colour fundus photography and fluorescein angiography (FA) were performed using a fundus camera system (TRC-50X/IMAGEnet; Topcon, Tokyo, Japan). Fluorescein angiograms were obtained after injection of 5 ml of a 10% sodium fluorescein solution. For macular ischaemia, the criteria adopted were those of the Early Treatment Diabetic Retinopathy Study (ETDRS)14 classification of the outline of the foveal avascular zone (FAZ).
Criteria for inclusion in the study were age 18 years or older, HR-PDR with or without DMO, and VA≥20/200. Exclusion criteria were pretreatment for DR (laser, intraocular medications and surgeries), pre-retinal and vitreous haemorrhage, presence of changes in the vitreous–retinal interface (epiretinal membrane, macular hole and vitreo–retinal traction syndrome), evidence of active external eye infection such as blepharitis, prior thromboembolic events, including myocardial infarction, stroke and deep vein thrombosis, systolic and diastolic blood pressures higher than 180 mm Hg and 110 mm Hg, respectively, glycated haemoglobin levels >15%, chronic renal failure, major surgery within 1 month and previous systemic antivascular endothelial growth factor.
Study design
The first eye was randomised to the study group (PRP with IVB injections) and the other eye to the control group (PRP) only. A schematic representation of the study protocol is shown in figure 1.
Study protocol. IVB, intravitreal bevacizumab; PRP, panretinal photocoagulation.
Treatment protocol
Patients received two IVB injections (1.25 mg/0.05 ml) with an interval of 3 weeks between injections. All injections were performed following international guidelines in operating rooms with sterile drapes and 5% topical povidone,15 and prophylactic topical antibiotics were administered 2 days before and 3 days after injection.
After evaluation, patients underwent laser episodes PRP in both eyes using the double frequency ND-YAG laser (532 nm) (Alcon, Ophthalas 532 EyeLite Laser Photocoagulator) through the 160 Mainster OMRA-PRP lens (Volk Instruments, Bullevue, California, USA).
The eye randomised to the study group received an IVB injection first followed by PRP after 1 week. The second IVB injection was administered at the end of the third episode.
PRP was performed weekly over 3 weeks, totalling three settings, starting from the inferior retina. The number of applications ranged between 300 and 500 shots (500 µm spot) per episode. The retinal laser applications were performed with the size of the sight set at 250 μm, an exposure time between 0.1 and 0.2 ms and moderate intensity (200–500 mW).16 If concomitant DMO existed, this was treated at the same time as PRP was performed, according to Olk et al17 and the ETDRS.18
Follow-up protocol
All patients were followed for 6 months, with returns in the first, third and sixth months. The ocular examination at all follow-up visits consisted of CS, intraocular pressure using a Goldmann applanation tonometer, slit lamp biomicroscopy, fundus examination, OCT and FA.
Main outcome measures
The primary outcome measures were changes in VCTS at 1, 3 and 6 months between the groups with and without DMO.
Secondary outcome measures were changes in VCTS within each group with and without DMO. Potential injection related complications, such as ocular hypertension, lens opacity progression and anterior chamber reaction, were studied for evaluation of ocular safety, and arterial thromboembolic events for systemic safety.
Statistical analysis
A Kolmogorov–Smirnov test was used to evaluate the normality assumption. Changes in CS between and within groups were assessed using the Wilcoxon signed rank test. The χ2 test was used to identify if divergence in patients with unilateral DMO occurred between groups. A p value <0.05 was considered statistically significant. Statistical analyses were performed using the software SPSS V.15.0 (SPSS Inc, Chicago, Illinois, USA).
Results
Patients at baseline
Between February 2011 and June 2012, about 600 referred patients with type 2 diabetes mellitus (DM) were evaluated and 42 patients (84 eyes) with HR-PDR were enrolled in the study. Mean age was 56 years (range 43–73) and 33.3% were women. The majority (66%) were users of insulin and 68.5% were phakic. Duration of DM and systemic arterial hypertension was 19 and 9 years, respectively, and mean glycated haemoglobin was 9.1. Baseline characteristics are shown in table 1.
Baseline systemic characteristics
During follow-up, seven patients presented vitreous haemorrhage and were removed from the study. Of the 35 remaining patients, 10 (28.6%) did not have DMO in either eye, 13 (37.2%) had DMO in one and 12 (34.2%) had DMO in both eyes. No significant difference was observed between groups regarding the presence or absence of unilateral DMO (p=0.811, χ2 test).
Both eyes of the same patients were randomised. Of the total of 35 patients, the mean best corrected visual acuity (logMAR) was 0.28 (range −0.01 to 1.00) in the study group (PRP+bevacizumab) and 0.24 (range −0.02 to 0.88) in the control group (PRP only). Mean FT was 305.50 μm (range 171–578 μm) in the control group and 295.5 μm (range 179–600 μm) in the study group (table 2).
Baseline ocular characteristics
In the 35 patients with and without DMO subjected to FA, the mean outline of FAZ was 2.7 (1.16) and 2.7 (0.92) degrees in the study and control groups, respectively. The macula of only two patients could not be classified. The mean number of burns was 1479 (229) and 1475 (239) in the study and control groups, respectively (p=0.75) (table 2).
Primary outcomes
A comparative analysis of CS in the 35 patients with and without DMO showed no significant between group differences at baseline in 1.5, 3, 6, 12 or 18 cpd (p=0.740, p=0.986, p=0.105, p=0.068 and p=0.782, respectively) (figure 2); 1 month, 1.5–18 cpd (p=0.252, p=0.731, p=1.000, p=0.987 and p=0.067, respectively) (figure 2); 3 months, 1.5–18 cpd (p=0.807, p=0.343, p=0.708, p=0.990 and p=0.869, respectively) (figure 2); and 6 months of follow-up, 1.5–18 cpd (p=0.807, p=0.564, p=0.562, p=0.767 and p=0.542, respectively) (figure 2).
Comparison of mean contrast sensitivity threshold scores in patients with and without macular oedema, in the control group and in the study group. (A) Baseline; (B) 1 month; (C) 3 months; (D) 6 months of follow-up.
When only patients with DMO (FT≥250 μm) were considered, there was also no statistically significant difference between groups (with the exception of spatial frequency of 12 cpd at baseline) at baseline 1.5–18 cpd (p=1.000, p=0.672, p=0.128, p=0.026 and p=0.686, respectively); 1 month, 1.5–18 cpd (p=0.058, p=0.572, p=0.256, p=0.528 and p=0.373, respectively); 3 months, 1.5–18 cpd (p=0.251, p=0.635, p=0.223, p=0.284 and p=0.959, respectively) and 6 months of follow-up, 1.5–18 (p=0.878, p=0.874, p=0.504, p=0.330 and p=0.441, respectively).
Secondary outcomes
When the results between moments in each group were compared separately, the control group had the following significance values for all 35 eyes: at 1 month, 1.5–18 cpd (p=0.047, p=0.903, p=0.111, p=0.025 and p=0.015, respectively); 3 months, 1.5–18 cpd (p=0.989, p=0.184, p=0.428, p=0.023 and p=0.903, respectively); and 6 months of follow-up, 1.5–18 cpd (p=0.122, p=0.691, p=0.004, p=0.005 and p=0.298, respectively) (figure 3). All significances above were deteriorations.
Comparison of mean contrast sensitivity threshold scores in patients with and without macular oedema, between baseline and follow-up times, in the control and study groups separately.
The study group had the following significance values: at 1 month, 1.5–18 cpd (p=0.321, p=0.539, p=0.972, p=0.741 and p=0.970, respectively); 3 months, 1.5–18 cpd (p=0.850, p=0.013 for better, p=0.210, p=0.442 and p=0.986, respectively); and 6 months of follow-up, 1.5–18 cpd (p=0.599, p=0.861, p=0.715, p=1.000 and p=0.262, respectively) (figure 3).
Only one statistically significant result was observed in patients with DMO in both eyes in the control group; the results at 1 month were, 1.5–18 cpd (p=0.169, p=0.778, p=0.344, p=0.233 and p=0.833, respectively); 3 months, 1.5–18 cpd (p=319, p=0.875, p=0.635, p=0.011 deteriorating and p=0.340, respectively); and 6 months of follow-up, 1.5–18 cpd (p=0.233, p=0.722, p=0.066, p=0.090 and p=0.600, respectively) (figure 3).
In the study group, there was no statistically significant results between moments; the results at 1 month were, 1.5–18 cpd (p=0.833, p=0.482, p=0.786, p=0.092 and p=0.833, respectively); 3 months, 1.5–18 cpd (p=1.000, p=0.674, p=0.673, p=0.276 and p=0.325, respectively); and 6 months of follow-up, 1.5–18 cpd (p=0.496, p=0.463, p=0.351, p=0.310, and p=0.887, respectively).
None of patients had complications such as ocular hypertension, lens opacity progression, anterior chamber reaction or arterial thromboembolic events.
Discussion
Even today, the complete effects of PRP on visual function are unknown, especially on CS, at different stages of DR. In the literature there are conflicting data regarding deterioration of CS in patients with PDR who were treated with PRP. Mackie and Walsh19 illustrated, in a cross sectional study, that PRP causes impairment in the mean threshold of CS. Khosla et al20 performed a prospective study and also demonstrated deterioration at the initial follow-up that was not maintained at the end of 3 months. However, these results were not confirmed by Canning et al21, irrespective of the wavelength of the light used, recruiting patients with good VA≥6/12 and performing PRP in only one session.
In the current study, when the groups were evaluated separately for each moment, we could also observe the effect of PRP on CS (studying the control group, subjected to PRP alone) for the treatment of HR-PDR. Our data showed deterioration in all spatial frequencies and at all stages of follow-up, especially in the 12 cpd measurement (figure 3), in agreement with previous studies that indicated a deleterious effect of PRP on CS function.19 ,20 Nevertheless, CS in the study group showed stabilisation during the entire 6 months of follow-up, and even showed improvement in spatial frequency 3 (p=0.013) cpd (figure 3). We consider that the most likely explanation is that bevacizumab may have acted by inhibiting the formation of MO by PRP and improving MO in those who already had it before treatment.9 This finding is in agreement with previous studies that showed improved VA and reduced MO in patients subjected to PRP combined with bevacizumab compared with PRP alone.12
Although a difference in results was observed in the within group comparisons (our secondary outcome) favouring the group that received bevacizumab, and despite the fact that the sample size of this study was one of the largest compared with previously published studies evaluating the adjuvant effect of IVB and PRP,10–12 this clinical trial did not show a statistically significant difference in relation to CS between groups in the total sample when the two groups were compared at each moment in time after treatment.
We believe the absence of a statistically significant difference in such comparisons was probably due to the sample size, because when the groups were separated and their evolution in time during follow-up compared, the results clearly pointed in favour of the study group (figure 3). Although not reaching statistical significance, a clear tendency for improving CS was demonstrated in the study group compared with the control group (figure 2). The proportion of eyes without DMO (33 (47%)) was another possible explanation for the absence of a significant difference between the groups. It is very likely that if a greater proportion of eyes had had MO the positive effect of IVB injection would have be better demonstrated. The result regarding CS in our study is similar to previously published studies evaluating the effect of treatment on VA, and showed no statistically significant difference in VA when using bevacizumab combined with PRP in patients without DMO.10 ,11 Finally, other factors that may have influenced the results are DMO duration, the integrity of the photoreceptor22 and macular ischaemia. It is important to consider that in the current study the FAZ could be accurately outlined in all but two (6%) patients and significant macular ischaemia was found in a large proportion of eyes. Arend et al23 published a study in which a statistically significant negative correlation existed between FAZ size and CS at high spatial frequencies. As there is a positive correlation between the size and outline of FAZ,24 it is possible that CS was also affected in this study as the majority of patients had significant macular ischaemia (table 2) with more than half of the FAZ destroyed.14
In conclusion, our study indicates that PRP deteriorates CS and adjuvant use of multiple IVB injections to PRP seems to be an alternative treatment for HR-PDR patients due to preservation of CS during the entire period of follow-up. Multicentre clinical trials are necessary to confirm this.
References
Footnotes
-
Collaborators Daniel Araujo Ferraz.
-
Contributors RCP: conception, design, analysis and interpretation of the data, and drafting the article. LMVR: conception, design, analysis and interpretation of the data, and drafting the article. MLRM: analysis and interpretation of the data, drafting the article and revising it critically for important intellectual content. MKC: analysis and interpretation of the data. DEP: revising the manuscript critically for important intellectual content and final approval of the version to be published. WYT: conception, design, analysis and interpretation of the data, revising the manuscript critically for important intellectual content and final approval of the version to be published
-
Funding This study was supported by the Sao Paulo Research Foundation (FAPESP) No 2009/08895-1.
-
Competing interests None.
-
Ethics approval The study was approved by the Comissão de Ética para Análise de Projetos de Pesquisa—CAPPesq No 0277/10 from the University of Sao Paulo
-
Provenance and peer review Not commissioned; externally peer reviewed.
-
Data sharing statement Data are available on request from the corresponding author.