Association of advanced glycation end products with diabetic retinopathy in type 2 diabetes mellitus

Lingwen Ying,Yun Shen,Yang Zhang,Yikun Wang,Yong Liu,Jun Yin,Yufei Wang,Jingrong Yin,Wei Zhu,Yuqian Bao,Jian Zhou
DOI: https://doi.org/10.1016/j.diabres.2021.108880
IF: 8.18
2021-07-01
Diabetes Research and Clinical Practice
Abstract:<h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Aims</h3><p>Advanced glycation end products (AGEs) were reported to be associated with diabetes development and diabetes related complications when accumulated in high levels. This study investigated the association between AGEs and diabetic retinopathy (DR).</p><h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Methods</h3><p>A total of 1,471 patients with type 2 diabetes were enrolled. Fundus radiography was used for DR measurement. AGEs were detected through non-invasive skin autofluorescence method.</p><h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Results</h3><p>Patients with more advanced DR showed a much higher AGEs, and the prevalence of DR (based on the severity) increase with ascending AGEs quartiles (all <em>P</em> for trend &lt; 0.001). The multivariable-adjusted odds ratios of any DR across AGEs quartiles were 1.00, 1.69 (95% confidence interval [<em>CI</em>] 1.16–2.47), 1.58 (95%<em>CI</em> 1.06–2.37) and 1.60 (95%<em>CI</em> 1.05–2.44) (<em>P</em> for trend = 0.044), respectively. Similar results were found in vision-threatening DR (VTDR) subgroup (<em>P</em> for trend = 0.009). When AGEs was considered as a continuous variable by using restricted cubic splines, a graded positive association of AGEs with the odds of any presence of DR was observed (<em>P</em> for trend &lt; 0.001). Further, we found that AGEs presented the similar predictive value for any DR with glycated hemoglobin A<sub>1c</sub> (HbA<sub>1c</sub>). When it comes to VTDR, AGEs showed a significantly higher efficacy in early screening than HbA<sub>1c</sub> (<em>P</em> = 0.002). With a cut-off point of 77.1, the sensitivity, specificity and area under the curve of AGEs were 90.0% (95%<em>CI</em> 76.3–97.2%), 49.4% (95%<em>CI</em> 46.8–52.0%), and 0.728 (95%<em>CI</em> 0.704–0.750), respectively.</p><h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Conclusion</h3><p>Non-invasive measured skin AGEs, associated with the prevalence of all stages of DR, might be a more suitable indicator than HbA<sub>1c</sub> for mimicking the poor prognosis of hyperglycemia.</p>
endocrinology & metabolism
What problem does this paper attempt to address?