Purpose

To determine clinical effectiveness, safety, and cost-effectiveness of subthreshold micropulse laser (SML), compared with standard laser (SL), for diabetic macular edema (DME) with central retinal thickness (CRT) < 400 μm.

Design

Pragmatic, multicenter, allocation-concealed, double-masked, randomized, noninferiority trial.

Participants

Adults with center-involved DME < 400 μm and best-corrected visual acuity (BCVA) of > 24 Early Treatment Diabetic Retinopathy Study (ETDRS) letters in one/both eyes.

Methods

Randomization 1:1 to 577 nm SML or SL treatment. Retreatments were allowed. Rescue with intravitreal anti–vascular endothelial growth factor therapies or steroids was permitted if 10 or more ETDRS letter loss occurred, CRT increased > 400 μm, or both.

Main Outcome Measures

Primary outcome was mean change in BCVA in the study eye at 24 months (noninferiority margin 5 ETDRS letters). Secondary outcomes were mean change from baseline to month 24 in binocular BCVA; CRT and mean deviation of Humphrey 10-2 visual field in the study eye; percentage meeting driving standards; EuroQoL EQ-5D-5L, 25-item National Eye Institute Visual Function Questionnaire (NEI-VFQ-25), and Vision and Quality of Life Index (VisQoL) scores; cost per quality-adjusted life-years (QALYs) gained; adverse effects; and number of laser and rescue treatments.

Results

The study recruited fully (n = 266); 87% of SML-treated and 86% of SL-treated patients had primary outcome data. Mean ± standard deviation BCVA change from baseline to month 24 was –2.43 ± 8.20 letters and –0.45 ± 6.72 letters in the SML and SL groups, respectively. Subthreshold micropulse laser therapy was deemed not only noninferior but also equivalent to SL therapy because the 95% confidence interval (CI; –3.9 to –0.04 letters) lay wholly within both upper and lower margins of the permitted maximum difference (5 ETDRS letters). No statistically significant difference was found in binocular BCVA (0.32 ETDRS letters; 95% CI, –0.99 to 1.64 ETDRS letters; P = 0.63); CRT (–0.64 μm; 95% CI, –14.25 to 12.98 μm; P = 0.93); mean deviation of the visual field (0.39 decibels (dB); 95% CI, –0.23 to 1.02 dB; P = 0.21); meeting driving standards (percentage point difference, 1.6%; 95% CI, –25.3% to 28.5%; P = 0.91); adverse effects (risk ratio, 0.28; 95% CI, 0.06–1.34; P = 0.11); rescue treatments (percentage point difference, –2.8%; 95% CI, –13.1% to 7.5%; P = 0.59); or EQ-5D, NEI-VFQ-25, or VisQoL scores. Number of laser treatments was higher in the SML group (0.48; 95% CI, 0.18–0.79; P = 0.002). Base-case analysis indicated no differences in costs or QALYs.

Conclusions

Subthreshold micropulse laser therapy was equivalent to SL therapy, requiring slightly higher laser treatments.

中文翻译：

---

糖尿病性黄斑水肿和二极管亚阈值微脉冲激光

目的

与标准激光 (SL) 相比，确定亚阈值微脉冲激光 (SML) 治疗视网膜中央厚度 (CRT) < 400 μm 的糖尿病性黄斑水肿 (DME) 的临床有效性、安全性和成本效益。

设计

实用、多中心、分配隐藏、双掩蔽、随机、非劣效性试验。

参加者

中心受累 DME < 400 μm 且单眼/双眼最佳矫正视力 (BCVA) > 24 个早期糖尿病视网膜病变研究 (ETDRS) 字母的成人。

方法

随机化 1:1 至 577 nm SML 或 SL 治疗。允许重新治疗。如果发生 10 个或更多 ETDRS 字母丢失，CRT 增加 > 400 μm，或两者兼而有之，则允许使用玻璃体内抗血管内皮生长因子疗法或类固醇进行抢救。

主要观察指标

主要结果是 24 个月时研究眼 BCVA 的平均变化（非劣效性界值 5 个 ETDRS 字母）。次要结果是双眼 BCVA 从基线到第 24 个月的平均变化；研究眼的 CRT 和 Humphrey 10-2 视野的平均偏差；达到驾驶标准的百分比；EuroQoL EQ-5D-5L、25 项国家眼科研究所视觉功能问卷 (NEI-VFQ-25) 以及视力和生活质量指数 (VisQoL) 评分；获得的每质量调整生命年 (QALY) 的成本；不利影响; 以及激光和抢救治疗的次数。

结果

该研究完全招募（n = 266）；87% 的 SML 治疗患者和 86% 的 SL 治疗患者有主要结果数据。从基线到第 24 个月的平均 ± 标准差 BCVA 变化在 SML 和 SL 组中分别为 –2.43 ± 8.20 个字母和 –0.45 ± 6.72 个字母。亚阈值微脉冲激光治疗被认为不仅不劣于 SL 治疗，而且与 SL 治疗相当，因为 95% 置信区间（CI；–3.9 至 –0.04 个字母）完全位于允许的最大差异（5 个 ETDRS 字母）的上下边界内。双眼 BCVA 未发现统计学显着差异（0.32 ETDRS 字母；95% CI，–0.99 至 1.64 ETDRS 字母；P  = 0.63）；CRT（–0.64 μm；95% CI，–14.25 至 12.98 μm；P = 0.93); 视野的平均偏差（0.39 分贝 (dB)；95% CI，–0.23 至 1.02 dB；P  = 0.21）；符合驾驶标准（百分点差异，1.6%；95% CI，–25.3% 至 28.5%；P  = 0.91）；不良反应（风险比，0.28；95% CI，0.06–1.34；P  = 0.11）；救援治疗（百分点差异，–2.8%；95% CI，–13.1% 至 7.5%；P  = 0.59）；或 EQ-5D、NEI-VFQ-25 或 VisQoL 分数。SML 组的激光治疗次数较多（0.48；95% CI，0.18–0.79；P  = 0.002）。基本案例分析表明成本或 QALY 没有差异。

结论

亚阈值微脉冲激光治疗相当于 SL 治疗，需要稍高的激光治疗。