Mwanza Jean-Claude, Sayyad Fouad E, Aref Ahmad A, Budenz Donald L
Bascom Palmer Eye Institute, University of Miami School of Medicine, Miami, Florida, USA.
Ophthalmic Surg Lasers Imaging. 2012 Nov-Dec;43(6 Suppl):S67-74. doi: 10.3928/15428877-20121003-01.
To compare rates of abnormal peripapillary retinal nerve fiber layer (RNFL) and macular ganglion cell layer scans acquired with Cirrus HD-OCT (Carl Zeiss Meditec, Inc., Dublin, CA) and RTVue-100 (Optovue Inc., Fremont, CA) in healthy myopic eyes.
Forty-one non-glaucomatous myopic eyes (41 individuals) were scanned with Cirrus to measure RNFL and ganglion cell-inner plexiform layer (GCIPL) and with RTVue to measure peripapillary RNFL and ganglion cell complex (GCC) thicknesses. Rates of abnormal scans were calculated and compared between devices. Inter-device agreement in falsely classifying scans as abnormal was also assessed.
The rate of abnormal average and four-quadrant RNFL was 4.8% to 7.3% on Cirrus and 2.4% to 9.7% on RTVue (P > .05). The overall rate of abnormal scans was 19.2% on Cirrus and 29.3% on RTVue (P = .3). Rates of abnormal Cirrus average and segmental GCIPL (12.2% to 17%) were similar to those of RTVue average and segmental GCC (9.7% to 14.6%) (P > .05). The overall rate of abnormal GCIPL (36.6%) was higher than that of GCC (14.6%) (P = .023). The inter-device agreement was poor for average RNFL (κ = -0.09), very good for average ganglion cell (κ = 0.81), and fair for overall RNFL (κ = 0.35) and overall ganglion cell (κ = 0.34).
The high rates of abnormal RNFL and ganglion cell layer scans on both devices call for caution, particularly when attempting to diagnose glaucoma in myopic eyes using these devices. The RNFL and ganglion cell layer analyses may not be interchangeable on either of these devices. These two devices are not interchangeable for classifying healthy myopic eyes based on RNFL or ganglion cell layer analysis.
比较在健康近视眼中，使用Cirrus HD-OCT（卡尔蔡司医疗技术公司，加利福尼亚州都柏林）和RTVue-100（奥普托视公司，加利福尼亚州弗里蒙特）获取的视乳头周围视网膜神经纤维层（RNFL）和黄斑神经节细胞层扫描的异常率。
对41只非青光眼性近视眼睛（41名个体）使用Cirrus进行扫描以测量RNFL和神经节细胞-内丛状层（GCIPL），并使用RTVue测量视乳头周围RNFL和神经节细胞复合体（GCC）厚度。计算并比较两种设备的异常扫描率。还评估了两种设备在将扫描错误分类为异常方面的一致性。
Cirrus上平均和四个象限RNFL的异常率为4.8%至7.3%，RTVue上为2.4%至9.7%（P>.05）。Cirrus上异常扫描的总体率为19.2%，RTVue上为29.3%（P =.3）。Cirrus平均和节段性GCIPL的异常率（12.2%至17%）与RTVue平均和节段性GCC的异常率（9.7%至14.6%）相似（P>.05）。GCIPL的总体异常率（36.6%）高于GCC（14.6%）（P =.023）。两种设备在平均RNFL方面的一致性较差（κ = -0.09），在平均神经节细胞方面非常好（κ = 0.81），在总体RNFL方面一般（κ = 0.35），在总体神经节细胞方面一般（κ = 0.34）。
两种设备上RNFL和神经节细胞层扫描的高异常率需要谨慎对待，尤其是在使用这些设备试图诊断近视眼中的青光眼时。在这两种设备上，RNFL和神经节细胞层分析可能不可互换。基于RNFL或神经节细胞层分析对健康近视眼睛进行分类时，这两种设备不可互换。
