Mehta Hemal, Hennings Charles, Gillies Mark C, Nguyen Vuong, Campain Anna, Fraser-Bell Samantha
Royal Free London NHS Foundation Trust, London, UK.
Cochrane Database Syst Rev. 2018 Apr 18;4(4):CD011599. doi: 10.1002/14651858.CD011599.pub2.
The combination of steroid and anti-vascular endothelial growth factor (VEGF) intravitreal therapeutic agents could potentially have synergistic effects for treating diabetic macular oedema (DMO). On the one hand, if combined treatment is more effective than monotherapy, there would be significant implications for improving patient outcomes. Conversely, if there is no added benefit of combination therapy, then people could be potentially exposed to unnecessary local or systemic side effects.
To assess the effects of intravitreal agents that block vascular endothelial growth factor activity (anti-VEGF agents) plus intravitreal steroids versus monotherapy with macular laser, intravitreal steroids or intravitreal anti-VEGF agents for managing DMO.
We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (which contains the Cochrane Eyes and Vision Trials Register) (2018, Issue 1); Ovid MEDLINE; Ovid Embase; LILACS; the ISRCTN registry; ClinicalTrials.gov and the ICTRP. The date of the search was 21 February 2018.
We included randomised controlled trials (RCTs) of intravitreal anti-VEGF combined with intravitreal steroids versus intravitreal anti-VEGF alone, intravitreal steroids alone or macular laser alone for managing DMO. We included people with DMO of all ages and both sexes. We also included trials where both eyes from one participant received different treatments.
We used standard methodological procedures recommended by Cochrane.Two authors independently reviewed all the titles and abstracts identified from the electronic and manual searches against the inclusion criteria. Our primary outcome was change in best corrected visual acuity (BCVA) between baseline and one year. Secondary outcomes included change in central macular thickness (CMT), economic data and quality of life. We considered adverse effects including intraocular inflammation, raised intraocular pressure (IOP) and development of cataract.
There were eight RCTs (703 participants, 817 eyes) that met our inclusion criteria with only three studies reporting outcomes at one year. The studies took place in Iran (3), USA (2), Brazil (1), Czech Republic (1) and South Korea (1). Seven studies used the unlicensed anti-VEGF agent bevacizumab and one study used licensed ranibizumab. The study that used licensed ranibizumab had a unique design compared with the other studies in that included eyes had persisting DMO after anti-VEGF monotherapy and received three monthly doses of ranibizumab prior to allocation. The anti-VEGF agent was combined with intravitreal triamcinolone in six studies and with an intravitreal dexamethasone implant in two studies. The comparator group was anti-VEGF alone in all studies; two studies had an additional steroid monotherapy arm, another study had an additional macular laser photocoagulation arm. Whilst we judged these studies to be at low risk of bias for most domains, at least one domain was at unclear risk in all studies.When comparing anti-VEGF/steroid with anti-VEGF monotherapy as primary therapy for DMO, we found no meaningful clinical difference in change in BCVA (mean difference (MD) -2.29 visual acuity (VA) letters, 95% confidence interval (CI) -6.03 to 1.45; 3 RCTs; 188 eyes; low-certainty evidence) or change in CMT (MD 0.20 μm, 95% CI -37.14 to 37.53; 3 RCTs; 188 eyes; low-certainty evidence) at one year. There was very low-certainty evidence on intraocular inflammation from 8 studies, with one event in the anti-VEGF/steroid group (313 eyes) and two events in the anti-VEGF group (322 eyes). There was a greater risk of raised IOP (Peto odds ratio (OR) 8.13, 95% CI 4.67 to 14.16; 635 eyes; 8 RCTs; moderate-certainty evidence) and development of cataract (Peto OR 7.49, 95% CI 2.87 to 19.60; 635 eyes; 8 RCTs; moderate-certainty evidence) in eyes receiving anti-VEGF/steroid compared with anti-VEGF monotherapy. There was low-certainty evidence from one study of an increased risk of systemic adverse events in the anti-VEGF/steroid group compared with the anti-VEGF alone group (Peto OR 1.32, 95% CI 0.61 to 2.86; 103 eyes).One study compared anti-VEGF/steroid versus macular laser therapy. At one year investigators did not report a meaningful difference between the groups in change in BCVA (MD 4.00 VA letters 95% CI -2.70 to 10.70; 80 eyes; low-certainty evidence) or change in CMT (MD -16.00 μm, 95% CI -68.93 to 36.93; 80 eyes; low-certainty evidence). There was very low-certainty evidence suggesting an increased risk of cataract in the anti-VEGF/steroid group compared with the macular laser group (Peto OR 4.58, 95% 0.99 to 21.10, 100 eyes) and an increased risk of elevated IOP in the anti-VEGF/steroid group compared with the macular laser group (Peto OR 9.49, 95% CI 2.86 to 31.51; 100 eyes).One study provided very low-certainty evidence comparing anti-VEGF/steroid versus steroid monotherapy at one year. There was no evidence of a meaningful difference in BCVA between treatments at one year (MD 0 VA letters, 95% CI -6.1 to 6.1, low-certainty evidence). Likewise, there was no meaningful difference in the mean CMT at one year (MD - 9 μm, 95% CI -39.87μm to 21.87μm between the anti-VEGF/steroid group and the steroid group. There was very low-certainty evidence on raised IOP at one year comparing the anti-VEGF/steroid versus steroid groups (Peto OR 0.75, 95% CI 0.16 to 3.55).No included study reported impact of treatment on patients' quality of life or economic data. None of the studies reported any cases of endophthalmitis.
AUTHORS' CONCLUSIONS: Combination of intravitreal anti-VEGF plus intravitreal steroids does not appear to offer additional visual benefit compared with monotherapy for DMO; at present the evidence for this is of low-certainty. There was an increased rate of cataract development and raised intraocular pressure in eyes treated with anti-VEGF plus steroid versus anti-VEGF alone. Patients were exposed to potential side effects of both these agents without reported additional benefit. The majority of the evidence comes from studies of bevacizumab and triamcinolone used as primary therapy for DMO. There is limited evidence from studies using licensed intravitreal anti-VEGF agents plus licensed intravitreal steroid implants with at least one year follow-up. It is not known whether treatment response is different in eyes that are phakic and pseudophakic at baseline.
玻璃体内注射类固醇与抗血管内皮生长因子（VEGF）治疗药物联合使用，可能对治疗糖尿病性黄斑水肿（DMO）具有协同作用。一方面，如果联合治疗比单一疗法更有效，那么对改善患者预后将具有重要意义。相反，如果联合治疗没有额外益处，那么人们可能会面临不必要的局部或全身副作用。
评估玻璃体内注射阻断血管内皮生长因子活性的药物（抗VEGF药物）加玻璃体内注射类固醇与黄斑激光、玻璃体内注射类固醇或玻璃体内注射抗VEGF药物单一疗法治疗DMO的效果。
我们检索了Cochrane对照试验中心注册库（CENTRAL）（其中包含Cochrane眼科和视力试验注册库）（2018年第1期）；Ovid MEDLINE；Ovid Embase；LILACS；国际标准随机对照试验编号注册库；ClinicalTrials.gov和国际临床试验注册平台。检索日期为2018年2月21日。
我们纳入了玻璃体内注射抗VEGF联合玻璃体内注射类固醇与单纯玻璃体内注射抗VEGF、单纯玻璃体内注射类固醇或单纯黄斑激光治疗DMO的随机对照试验（RCT）。我们纳入了所有年龄和性别的DMO患者。我们还纳入了同一参与者的双眼接受不同治疗的试验。
我们采用了Cochrane推荐的标准方法程序。两位作者独立审查了通过电子和手动检索确定的所有标题和摘要，以符合纳入标准。我们的主要结局是基线至一年期间最佳矫正视力（BCVA）的变化。次要结局包括中心黄斑厚度（CMT）的变化、经济数据和生活质量。我们考虑了不良反应，包括眼内炎症、眼压升高（IOP）和白内障的发生。
有8项RCT（703名参与者，817只眼）符合我们的纳入标准，只有3项研究报告了一年时的结局。这些研究分别在伊朗（3项）、美国（2项）、巴西（1项）、捷克共和国（1项）和韩国（1项）进行。7项研究使用了未获许可的抗VEGF药物贝伐单抗，1项研究使用了获许可的雷珠单抗。与其他研究相比，使用获许可雷珠单抗的研究设计独特，即纳入的眼睛在抗VEGF单一疗法后仍存在DMO，并在分配前接受了三个月一次的雷珠单抗剂量。6项研究中抗VEGF药物与玻璃体内注射曲安奈德联合使用，2项研究中与玻璃体内地塞米松植入物联合使用。所有研究中的对照组均为单纯抗VEGF；2项研究有额外的类固醇单一疗法组，另一项研究有额外的黄斑激光光凝组。虽然我们判断这些研究在大多数领域的偏倚风险较低，但所有研究中至少有一个领域的风险尚不清楚。当比较抗VEGF/类固醇与抗VEGF单一疗法作为DMO的主要治疗方法时，我们发现一年时BCVA的变化（平均差（MD）-2.29视力（VA）字母，95%置信区间（CI）-6.03至1.45；3项RCT；188只眼；低确定性证据）或CMT的变化（MD 0.20μm，95%CI -37.14至37.53；3项RCT；188只眼；低确定性证据）没有有意义的临床差异。8项研究关于眼内炎症的证据非常低确定性，抗VEGF/类固醇组（313只眼）有1例事件，抗VEGF组（322只眼）有2例事件。与抗VEGF单一疗法相比，接受抗VEGF/类固醇治疗的眼睛眼压升高的风险更大（Peto比值比（OR）8.13，95%CI 4.67至14.16；635只眼；8项RCT；中等确定性证据），白内障发生的风险更大（Peto OR 7.49，95%CI 2.87至19.60；635只眼；8项RCT；中等确定性证据）。一项研究提供的低确定性证据表明，与单纯抗VEGF组相比，抗VEGF/类固醇组全身不良事件的风险增加（Peto OR 1.32，95%CI 0.61至2.86；103只眼）。一项研究比较了抗VEGF/类固醇与黄斑激光治疗。一年时，研究者未报告两组在BCVA变化（MD 4.00 VA字母，95%CI -2.70至10.70；80只眼；低确定性证据）或CMT变化（MD -16.00μm，95%CI -68.93至36.93；80只眼；低确定性证据）方面有有意义的差异。非常低确定性的证据表明，与黄斑激光组相比，抗VEGF/类固醇组白内障的风险增加（Peto OR 4.58，95% 0.99至21.10，100只眼），与黄斑激光组相比，抗VEGF/类固醇组眼压升高的风险增加（Peto OR 9.49，95%CI 2.86至31.51；100只眼）。一项研究提供了非常低确定性的证据，比较了抗VEGF/类固醇与类固醇单一疗法一年时的情况。一年时治疗之间在BCVA方面没有有意义差异的证据（MD 0 VA字母，95%CI -6.1至6.1，低确定性证据）。同样，抗VEGF/类固醇组与类固醇组在一年时平均CMT方面也没有有意义的差异（MD - 9μm，95%CI -39.87μm至21.87μm）。比较抗VEGF/类固醇与类固醇组一年时眼压升高的证据非常低确定性（Peto OR 0.75，95%CI 0.16至3.55）。纳入的研究均未报告治疗对患者生活质量或经济数据的影响。没有研究报告任何眼内炎病例。
与DMO单一疗法相比，玻璃体内注射抗VEGF加玻璃体内注射类固醇似乎没有提供额外的视力益处；目前关于这方面的证据确定性较低。与单纯抗VEGF相比，抗VEGF加类固醇治疗的眼睛白内障发生率和眼压升高率增加。患者在没有额外益处报告的情况下暴露于这两种药物的潜在副作用中。大多数证据来自使用贝伐单抗和曲安奈德作为DMO主要治疗方法的研究。使用获许可的玻璃体内抗VEGF药物加获许可的玻璃体内类固醇植入物且至少随访一年的研究证据有限。尚不清楚基线时为晶状体眼和人工晶状体眼的眼睛治疗反应是否不同。
