Highlight
- Continuous glucose monitoring (CGM) was associated with significantly reduced all-cause mortality among U.S. veterans receiving dialysis with diabetes.
- Propensity score matching and adjustment for confounders confirmed robust survival benefits for incident CGM users compared to non-users.
- CGM offers comprehensive glycemic data beyond traditional markers, addressing limitations in dialysis patients where standard glucose monitoring is challenging.
- Further research is needed to elucidate mechanisms by which CGM influences outcomes and to evaluate broader impacts on dialysis-related complications.
Background: Diabetes Management Challenges in Dialysis Patients
Patients with end-stage kidney disease (ESKD) undergoing dialysis represent a population with markedly high cardiovascular risk and mortality. When diabetes coexists, glycemic management becomes critically important yet challenging. Traditional measures of glycemic control, such as hemoglobin A1c (HbA1c), are less accurate in dialysis patients due to altered red blood cell lifespan and other metabolic changes, complicating risk assessment and therapeutic decisions. Episodic fingerstick glucose monitoring is cumbersome and often fails to capture glycemic variability, including asymptomatic hypo- or hyperglycemia that may contribute to adverse cardiovascular and mortality outcomes.
Continuous glucose monitoring (CGM) technology provides automated, minimally invasive, real-time data on interstitial glucose levels, capturing detailed patterns and fluctuations. While CGM has demonstrated clinical utility in general diabetes populations, its impact on clinical outcomes in patients receiving dialysis remains uncertain. This study aimed to evaluate whether CGM use is associated with improved survival outcomes among U.S. veterans receiving dialysis who have diabetes.
Study Design and Methods
This observational cohort study utilized linked data from the Veterans Affairs (VA), the U.S. Renal Data System (USRDS), and Medicare databases, covering January 2012 through December 2023. The study population included veterans with diagnosed diabetes undergoing maintenance dialysis. The exposure of interest was incident CGM use, defined as new CGM prescription and initiation, compared to non-CGM users.
To control for confounding by indication—a key issue in observational designs—propensity score (PS) matching was employed to balance baseline characteristics between CGM users and non-users. Cox proportional hazards models were used to estimate hazard ratios (HRs) for all-cause mortality, implementing both unadjusted and doubly adjusted models (accounting for residual confounding). Two analytic cohorts were assessed: a complete case cohort of 2,008 patients with no missing data and a larger cohort of 3,088 patients analyzed with multiple imputation to handle missing covariates. Survival was tracked through February 2025.
Key Findings
Among the fully PS-matched groups, CGM use was consistently associated with statistically significant reductions in mortality risk compared to non-CGM use. In the complete case cohort (n=2,008), CGM users had a 14% lower mortality risk in unadjusted analysis (HR 0.86; 95% CI 0.76 to 0.98) and a 17% lower risk after multivariable adjustment (HR 0.83; 95% CI 0.75 to 0.92).
Similarly, in the multiple imputation cohort (n=3,088), the survival advantage persisted, with unadjusted HR 0.88 (95% CI 0.77 to 0.99) and adjusted HR 0.84 (95% CI 0.73 to 0.96). These effect sizes indicate a consistent association between CGM use and improved survival in this high-risk population.
The study did not report specific safety outcomes directly related to CGM use, but the observational design and comprehensive data linkage provide strong real-world evidence supporting the clinical relevance of CGM among dialysis patients with diabetes.
Expert Commentary
This investigation offers important insights into the potential benefits of CGM in a population where glycemic management is notoriously complex and prognostically important. The robust methodological approach, including PS matching and multiple imputation for missing data, strengthens the validity of findings despite the inherent limitations of observational studies.
Mechanistically, CGM may improve survival by enabling tighter and safer glycemic control through better detection of hypoglycemia and avoidance of extreme glucose excursions, which are linked to cardiovascular morbidity and mortality. Additionally, CGM may enhance patient engagement and provider decision-making by furnishing detailed glycemic profiles that surpass isolated glucose checks or HbA1c.
However, the study cannot establish causality, and residual confounding cannot be excluded. The veteran cohort predominantly includes older males with specific socio-demographic features, and hence, generalizability to non-veteran populations or women may be limited. Future randomized controlled trials or pragmatic intervention studies are warranted to confirm these observational findings, clarify underlying mechanisms, and explore impacts on other clinical outcomes such as hospitalization, cardiovascular events, and quality of life.
Conclusion
Incident use of continuous glucose monitoring among U.S. veterans receiving dialysis with diabetes was associated with a significant survival benefit compared to non-CGM use. CGM offers a promising tool to overcome conventional challenges in glycemic monitoring in this vulnerable group. This evidence supports broader clinical consideration of CGM integration into diabetes care protocols for dialysis patients. Continued research should aim at understanding mechanistic pathways, cost-effectiveness, and optimal implementation strategies in varied clinical settings.
Funding and ClinicalTrials.gov
Information on funding sources was not specified in the source article. The study design, a retrospective cohort analysis based on registry data, does not appear registered under ClinicalTrials.gov.
References
1. Narasaki Y, Kalantar-Zadeh K, Zisman-Ilani Y, et al. Continuous Glucose Monitoring and Mortality Risk Among U.S. Veterans Receiving Dialysis With Diabetes. Diabetes Care. 2026 Jul;49(7):1285-1293. doi:10.2337/dc25-1678. PMID: 42224118.
2. Kovesdy CP, et al. Glycemic Control in Dialysis Patients: Pitfalls and Promises. Semin Dial. 2017 Sep-Oct;30(5):426-433.
3. Klonoff DC, et al. Clinical Utility and Impact of Continuous Glucose Monitoring in Diabetes Care. Diabetes Technol Ther. 2020 Feb;22(S2):S-38-S-45.
4. Kalantar-Zadeh K, et al. Comprehensive Review of Glucose Management in Dialysis Patients. Am J Kidney Dis. 2022 Mar;79(3):363-375.
