PDF(Pubmed)
Abstract：
UNASSIGNED: Diabetes is rapidly becoming a major cause of blindness among Kenyans, with the prevalence of any form of diabetes retinopathy (DR) ranging from 36% to 41%. Globally DR leads as a cause of vision loss in working age adults. In Kenya, specialized examinations are only available at national and some county referral hospitals through retina specialists, ophthalmologists or trained technicians. Thus, low coverage of retinal assessment and inadequate access to this service. An innovative DR fundus camera screening service run by ophthalmic nurses (ONs), ophthalmic clinical officers (OCOs) and county ophthalmologists was established since 2018.
UNASSIGNED: The purpose of this study was to investigate the diagnostic accuracy of DR digital retinal camera screening by ONs, OCOs and county ophthalmologist against that of a retina specialist measured by sensitivity and specificity as the primary outcomes.
UNASSIGNED: Cross sectional study conducted at 2 referral hospitals in Kenya. Using a Canon CR-2AF digital retinal camera patients with diabetes had a standard single shot of 45 degree view of the retina captured as image in each eye. This was graded for DR using the International Clinical Diabetic Retinopathy (ICDR) severity scale. All photos taken by the first graders (ON/OCO) were later assessed by the county hospital ophthalmologist who was blinded to their readings. The third grader (retina specialist) similarly was blinded to the readings of the first and second graders and assessed all the images from the 2 hospitals also using ICDR.
UNASSIGNED: A total of 308 patients with diabetes (median age 58 IQR 56-60, 53% female) were enrolled in the study. Sensitivity to identify any DR was (81.3%, 80.6%, and 81.54% for the OCO, ON and county ophthalmologist respectively). The corresponding specificities were 92.7%, 92.8% and 92.59%. Analysis of diagnostic accuracy of non-sight threatening DR against sight threatening DR revealed lower sensitivity for the three cadre groups although specificity remained high.
UNASSIGNED: In this study, ON and OCO with basic training in DR screening and photo grading performed screening of DR with high specificity. However, the sensitivity to detect sight threatening DR was generally low by all the cadres which may leave severe forms of DR undetected.
UNASSIGNED: The purpose of this study was to investigate the diagnostic accuracy of DR digital retinal camera screening by ONs, OCOs and county ophthalmologist against that of a retina specialist measured by sensitivity and specificity as the primary outcomes.
UNASSIGNED: Cross sectional study conducted at 2 referral hospitals in Kenya. Using a Canon CR-2AF digital retinal camera patients with diabetes had a standard single shot of 45 degree view of the retina captured as image in each eye. This was graded for DR using the International Clinical Diabetic Retinopathy (ICDR) severity scale. All photos taken by the first graders (ON/OCO) were later assessed by the county hospital ophthalmologist who was blinded to their readings. The third grader (retina specialist) similarly was blinded to the readings of the first and second graders and assessed all the images from the 2 hospitals also using ICDR.
UNASSIGNED: A total of 308 patients with diabetes (median age 58 IQR 56-60, 53% female) were enrolled in the study. Sensitivity to identify any DR was (81.3%, 80.6%, and 81.54% for the OCO, ON and county ophthalmologist respectively). The corresponding specificities were 92.7%, 92.8% and 92.59%. Analysis of diagnostic accuracy of non-sight threatening DR against sight threatening DR revealed lower sensitivity for the three cadre groups although specificity remained high.
UNASSIGNED: In this study, ON and OCO with basic training in DR screening and photo grading performed screening of DR with high specificity. However, the sensitivity to detect sight threatening DR was generally low by all the cadres which may leave severe forms of DR undetected.
摘要：
■糖尿病正迅速成为肯尼亚人失明的主要原因，任何形式的糖尿病视网膜病变（DR）的患病率从36％到41％不等。在全球范围内，DR是导致工作年龄成年人视力丧失的原因。在肯尼亚,专门检查仅在国家和一些县转诊医院通过视网膜专家进行，眼科医生或训练有素的技术人员。因此，视网膜评估覆盖率低，无法充分获得这项服务。由眼科护士(ON)运营的创新DR眼底照相机筛查服务，眼科临床干事（OCOs）和县眼科医生自2018年成立。
■这项研究的目的是探讨通过ON进行DR数字视网膜照相机筛查的诊断准确性，OCOs和县眼科医生对视网膜专家的敏感性和特异性作为主要结果。
■在肯尼亚的2家转诊医院进行的横断面研究。使用佳能CR-2AF数字视网膜照相机,患有糖尿病的患者具有作为每只眼睛中的图像捕获的视网膜的45度视图的标准单次拍摄。使用国际临床糖尿病视网膜病变（ICDR）严重程度量表对DR进行分级。一年级学生（ON/OCO）拍摄的所有照片后来都由县医院眼科医生评估，他对他们的读数视而不见。同样，三年级学生（视网膜专家）对一年级和二年级学生的读数视而不见，并使用ICDR评估了2家医院的所有图像。
■共308名糖尿病患者（中位年龄58IQR56-60，53%为女性）纳入研究。识别任何DR的敏感性为（81.3%，80.6%,OCO为81.54%，分别为ON和县眼科医生)。相应的特异性为92.7%，92.8%和92.59%。对无视力威胁DR对视力威胁DR的诊断准确性的分析显示，尽管特异性仍然很高，但三个干部组的敏感性较低。
■在这项研究中，ON和OCO在DR筛查和照片分级的基础培训下进行了高特异性的DR筛查。然而,所有干部对威胁视力的DR的敏感性普遍较低，这可能会导致严重的DR未被发现。
■这项研究的目的是探讨通过ON进行DR数字视网膜照相机筛查的诊断准确性，OCOs和县眼科医生对视网膜专家的敏感性和特异性作为主要结果。
■在肯尼亚的2家转诊医院进行的横断面研究。使用佳能CR-2AF数字视网膜照相机,患有糖尿病的患者具有作为每只眼睛中的图像捕获的视网膜的45度视图的标准单次拍摄。使用国际临床糖尿病视网膜病变（ICDR）严重程度量表对DR进行分级。一年级学生（ON/OCO）拍摄的所有照片后来都由县医院眼科医生评估，他对他们的读数视而不见。同样，三年级学生（视网膜专家）对一年级和二年级学生的读数视而不见，并使用ICDR评估了2家医院的所有图像。
■共308名糖尿病患者（中位年龄58IQR56-60，53%为女性）纳入研究。识别任何DR的敏感性为（81.3%，80.6%,OCO为81.54%，分别为ON和县眼科医生)。相应的特异性为92.7%，92.8%和92.59%。对无视力威胁DR对视力威胁DR的诊断准确性的分析显示，尽管特异性仍然很高，但三个干部组的敏感性较低。
■在这项研究中，ON和OCO在DR筛查和照片分级的基础培训下进行了高特异性的DR筛查。然而,所有干部对威胁视力的DR的敏感性普遍较低，这可能会导致严重的DR未被发现。
