Long-term variability of glycemic markers and risk of all-cause mortality in type 2 diabetes: A systematic review and meta-analysis
Year: 2025
Type: review
Study designs: Systematic Review, Meta-Analysis
Abstract: Background & aims This research aimed to assess how long-term fluctuations in glycemic indicators (FPG and HbA1c), measured by the coefficient of variation (CV), influence the risk of all-cause mortality in people with type 2 diabetes. Methods A review of all literature related to the variability of glycemic indicators published up to June 2025 was conducted using PubMed, Scopus, and Google Scholar. A total of 3,663 articles were found and evaluated based on the title, introduction, and full text, leading to the selection of 15 studies for the meta-analysis. Meta-analysis was performed using a random-effects model. Subgroup and meta-regression analyses were used to explore sources of heterogeneity. Results Fifteen studies ( n = 584,237) were included. Higher glycemic variability was generally linked to increased mortality risk, but this connection showed extreme statistical heterogeneity (I² >85% for both HbA1c-CV and FPG-CV). Therefore, single pooled estimates (e.g., HR: 0.39 for HbA1c-CV) were unreliable and required stratified interpretation; initial inverse associations were corrected after harmonizing effect coding (Table S1). Importantly, meta-regression revealed that cohort mean age was the strongest moderator, explaining nearly 48.3% (R² = 48.3%) of the between-study heterogeneity. The prognostic link was notably stronger in younger populations, suggesting a diminished effect in older adults due to competing risks. Sensitivity analyses confirmed the consistency of the overall effect direction. Conclusions Long-term glycemic variability is linked to all-cause mortality in type 2 diabetes. Given the extreme heterogeneity, our findings emphasize that age strongly modifies the risk estimates, suggesting that GV is a more potent prognostic marker in younger T2DM populations. Clinically, potential monitoring of GV should be highly individualized and considered hypothesis-generating, particularly in high-risk, younger subgroups. Standardization of metrics and further prospective studies are warranted.
Authors: Saba Rismantab Sani, Hamed Taheri, Amirhesam Amirbeik, Parisa alsadat Dadkhah, Fatemeh Chichagi, Anahita Kalanimehr, Abbas Pakmehr, Seyyed-Ghavam Shafagh, Roja Valipoor, Mahdi Nakhaee, Mahsa Aslani Khiavi, Kiarash Sadat Rafiei, Reza Khademi, Arshia Shafiei, Reza Hossein Zadeh, Mohadeseh Poudineh, Mahdyieh Naziri, Yaser Khakpour, Sourena Mohammad Hashem, Niloofar Deravi
Institutions: Guilan University of Medical Sciences, Kazan Federal University, Mashhad University of Medical Sciences, Isfahan University of Medical Sciences, University of Tehran, Tehran Heart Center, Tehran University of Medical Sciences, Qazvin University of Medical Sciences, Shiraz University of Medical Sciences, Iran University of Medical Sciences, Islamic Azad University, Tehran, Shahid Beheshti University, Shahid Beheshti University of Medical Sciences, Zanjan University of Medical Sciences
Language: English
Subfield: Endocrinology, Diabetes and Metabolism
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W4416934032