Scholarlink

Aim <p>Time in Tight Range (TITR) is physiologically closer to normoglycemia than time in range (TIR) and has been linked to microvascular risk. We aimed to evaluate associations of TITR with diabetic retinopathy (DR) and optical coherence tomography angiography (OCTA)-derived retinal parameters in adults with type 1 diabetes (T1D) and to compare its performance with TIR.</p> Methods <p>We retrospectively included adults with T1D who underwent continuous glucose monitoring (CGM) and OCTA. TITR–DR associations were tested using multivariable logistic regression with restricted cubic splines. Linear mixed-effects models assessed TITR/TIR associations with OCTA-derived parameters, followed by exploratory threshold-based comparisons. DR discrimination of TITR versus TIR was assessed using ROC analysis.</p> Results <p>We analyzed 259,568 glucose readings from 153 participants. Each 10-percentage-point increase in TITR was associated with 29.5% lower odds of DR (adjusted OR 0.705; <i>p</i> = 0.004), without evidence of nonlinearity. In continuous analyses, both TITR and TIR were positively associated with deep vascular complex (DVC) vessel density and perfusion area; in exploratory threshold-based analyses, the TITR 50% cutoff showed significant DVC differences, whereas the TIR 70% cutoff showed no consistent differences. DR discrimination was similar for TITR and TIR (<i>p</i> = 0.635), whereas TITR was more closely related to hypoglycemia exposure (<i>p</i> &lt; 0.01).</p> Conclusions <p>Higher TITR was associated with lower odds of DR and more favorable DVC vessel density and perfusion area in hospitalized adults with T1D. TITR showed DR discrimination comparable to TIR but may provide complementary information on hypoglycemia exposure and DVC differences in threshold-based analyses.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Time in tight range and time in range in relation to diabetic retinopathy and OCTA-derived retinal microvascular parameters in hospitalized adults with type 1 diabetes

  • 期刊论文

  • 发表日期：

    2026年8月3日

查看全文
  • Ruoxuan Zheng,
  • Cong Zhang,
  • Xiaotong Sun,
  • Jingjing Li,
  • Li Wang,
  • Shimeng Huang,
  • Rui Hou,
  • Siyu Dong,
  • Shuang Wu,
  • Hongxue Li,
  • Linlin Zhang,
  • Binglin Kuang,
  • Lei Zheng,
  • Baiwen Zhang,
  • Changwei Yang,
  • Hongyu Kuang

摘要

Aim

Time in Tight Range (TITR) is physiologically closer to normoglycemia than time in range (TIR) and has been linked to microvascular risk. We aimed to evaluate associations of TITR with diabetic retinopathy (DR) and optical coherence tomography angiography (OCTA)-derived retinal parameters in adults with type 1 diabetes (T1D) and to compare its performance with TIR.

Methods

We retrospectively included adults with T1D who underwent continuous glucose monitoring (CGM) and OCTA. TITR–DR associations were tested using multivariable logistic regression with restricted cubic splines. Linear mixed-effects models assessed TITR/TIR associations with OCTA-derived parameters, followed by exploratory threshold-based comparisons. DR discrimination of TITR versus TIR was assessed using ROC analysis.

Results

We analyzed 259,568 glucose readings from 153 participants. Each 10-percentage-point increase in TITR was associated with 29.5% lower odds of DR (adjusted OR 0.705; p = 0.004), without evidence of nonlinearity. In continuous analyses, both TITR and TIR were positively associated with deep vascular complex (DVC) vessel density and perfusion area; in exploratory threshold-based analyses, the TITR 50% cutoff showed significant DVC differences, whereas the TIR 70% cutoff showed no consistent differences. DR discrimination was similar for TITR and TIR (p = 0.635), whereas TITR was more closely related to hypoglycemia exposure (p < 0.01).

Conclusions

Higher TITR was associated with lower odds of DR and more favorable DVC vessel density and perfusion area in hospitalized adults with T1D. TITR showed DR discrimination comparable to TIR but may provide complementary information on hypoglycemia exposure and DVC differences in threshold-based analyses.