Increased use of continuous glucose monitoring in the United States has improved management of patients with type 1 and type 2 diabetes. Continuous glucose monitoring is also being promoted as a biofeedback wellness tool for healthy patients without diabetes. Despite its widespread use, it remains unclear how and whether patients will benefit continuous glucose monitoring. In this study, initiation of continuous glucose monitoring was associated with a reduced risk of subsequent development of DME, PDR, VH, and TRD in patients with NPDR when compared to a matched cohort of patients who were not initiated with continuous glucose monitoring. Furthermore, patients with NPDR initiated with continuous glucose monitoring had a lower risk of requiring intravitreal anti-VEGF injections, PRP, and PPV when compared to a matched cohort of patients not initiated with continuous glucose monitoring. These findings were also observed after adjusting for HBA1C levels and were consistent across subgroup analyses, regardless of age, gender, or insulin use. The findings of the current study suggest that additional glucose factors assayed by continuous glucose monitoring, including glucose variability and range of time, may affect the risk of microvascular changes associated with type 1 and type 2 diabetes.
In clinical practice, blood glucose status is assessed primarily by HBA1C, measuring the average blood glucose over a period of three months. Clinical trials also use HBA1C measurements to assess the risk of diabetes complications3,21 Despite extensive use, limitations of the HBA1C test exist, such as incomplete expression of blood glucose variability and incomplete expression of time in range. Independently of HBA1C, increased glucose variability and reduced time of range have been established as risk factors for type 2-related microvascular complications, including diabetes type 1 and DR. In a large single-center cohort study, Lu et al. We demonstrated that DR development was inversely correlated with higher times within the range of type 2 diabetes patients (Pearson correlation coefficient = −0.147; p < .001), independent of HBA1C level 10. Furthermore, blood glucose variability is recognized as an independent predictor of microvascular complications in type 1 and type 2 diabetes 24,25 by inducing oxidative stress and endothelial dysfunction. Given the growing literature suggesting the limitations of HBA1C, and the importance of time of glucose variability and range in predicting diabetes complications, innovative ways to assess glucose status have emerged.
Point-of-care blood glucose monitoring (BGM) and continuous glucose monitoring are two commonly used methods to assess glucose control in patients with type 1 and type 2 diabetes. BGM with point-of-care capillary or finger bar devices is essential for the management of patients with severe insulin-deficient diabetes, requiring frequent and burdensome capillary or finger mud devices, but may enhance the use of continuous glucose monitoring. Increases the range time, reduces the variability of glucose levels, the incidence of hypoglycemic events, and continuous glucose monitoring, a diabetic ketosidosis, provides unique protection against diabetic complications11,14,16,24,25. Despite the growing literature to support the use of continuous glucose monitoring in type 1 and type 2 diabetes, data are limited to quantifying its benefits in preventing diabetic complications27,28. Sartore et al. The standard deviation of the rate of change in blood glucose calculated every 2 hours during the monitoring period and the increase in continuous overlap net glucose effects were both aimed at estimating blood glucose variability using continuous glucose monitoring, and was apparently associated with increased DR in patients with type 1 and type 2 diabetes. [OR]1.03; p = .01 and or 1.02; p = .04) 29, respectively. The importance of this association was lost after correction for the presence of hypertension and hypercholesterolemia. Furthermore, Liu et al. The use of continuous glucose monitoring, a proxy for lower glycemic variability, showed that in the absence of continuous glucose monitoring among privately insured type 1 diabetes patients, was 14 with lower odds of DR and PDR and primarily white races (OR, 0.52; P = .008 and OR, 0.42; P = .004). It remains unclear whether the results highlighted by Liu et al. Generalizing to all adult populations of type 1 and type 2 diabetes also increases the risk of diabetes VTC and can extrapolate to the need for ocular intervention in patients using continuous glucose monitoring compared to patients not using continuous glucose monitoring. Although we did not directly measure blood glucose variability in the present study, the use of continuous glucose monitoring in patients with NPDR was likely to result in a reduced variability in glucose levels and was associated with a reduced risk of subsequent development of DME (HR, 0.80; P <.001), PDR (HR, 0.70; P <.001), VH (HR, 0.52; P <.001), and PDR (HR, 0.70; P <.001). 0.39; P = .003) Large heterogeneous real-world database of type 2 and NPDR matched patients over two years.
This study has several limitations that justify the consideration. As a retrospective analysis of a large set of disagreement-comprehensive medical record data, unaddressed potential confounders may affect the outcomes of this study. For example, selection bias in patient access to ongoing glucose monitoring could affect outcomes, as this analysis did not control for socioeconomic status and lifestyle factors. Furthermore, compliance with the use of continuous glucose monitoring may reflect more compliant patients in other areas of care that cannot be readily measured by this type of study design. Medication compliance, healthy dietary intake, and exercise can all directly affect the incidence and progression of microvascular complications. However, the use of continuous glucose monitoring is becoming the standard of care for type 1 diabetes, and recent FDA approval for marketing STELO biosensors has enabled continuous glucose monitoring to patients with type 1 and type 2 diabetes without a prescription13. The advent of other continuous glucose monitoring systems and newer versions of accuracy and ease of use on the market has also expanded the use of continuous glucose monitoring in patients with type 1 and type 2 diabetes. Future studies may assess how the use of hybrid closure techniques or insulin pumps is associated with the risk of developing diabetic complications and the risk of undergoing ocular intervention among patients with type 1 and type 2 diabetes. Furthermore, it is unclear how often patients in our study's continuous glucose monitoring cohort use continuous glucose monitoring after enrollment. Because most continuous glucose monitoring lasts for 2 weeks, the repeated use of range stabilization following continuous glucose monitoring, blood glucose variability, and selection bias in motivated patients may explain improved ocular outcomes among patients in the continuous glucose monitoring cohort compared to the control cohort. Diabetes type and duration of disease were not included in the analysis, and differences between both cohorts can confuse the outcome. Second, this study relies on accurate ICD-10 diagnosis and coding, and results may be affected by inaccuracies in true coding for DME, PDR, VH, TRD, PPV, PRP, and intravitreal anti-VEGF injections. Due to the relatively low developmental rate of VTC and the need for ocular intervention among patients in each cohort of this study, our study included only 4% and 0% of patients in the continuous glucose monitoring and control cohort, respectively. Although this may affect the results of our study, 0% of patients with severe NPDR after propensity score matching were included. Finally, patients who received ocular intervention for treatment of NPDR prior to index date were excluded from each cohort, but treatment during the study period could affect the development of NPDR VTCs. However, despite increased risk of requiring intravitreal anti-VEGF drug therapy, PRP, and PPV during the study period, patients enrolled in the continuous glucose monitoring cohort developed VTC at an increased rate compared to matched patients who were not initiated with continuous glucose monitoring. In the analysis, no statistical assumptions and interaction tests were performed to compare baseline characteristics. Finally, despite relatively stable practice patterns over the past 20 years, differences in PDR management existed during the enrollment period, which could affect the results of the study. Despite its limitations, this study strengthens increased support for ongoing glucose monitoring in the co-management of diabetic patients with DR, providing valuable insights to patients and providers during clinical practice and counseling.
Using a large real-world patient data set, this study demonstrated a reduced risk of developing subsequent DME, PDR, VH, and TRD in patients with NPDR initiated with continuous glucose monitoring, when compared to a matched cohort of patients who were not initiated with continuous glucose monitoring, even when HBA1C levels were controlled between both cohorts. Those initiated with continuous glucose monitoring had a lower risk of requiring intravitreal anti-VEGF drug therapy, PRP, and PPV when compared to a matched cohort of patients who did not initiate with continuous glucose monitoring. Although the use of continuous glucose monitoring is rapidly increasing in patients with type 1 diabetes, the results of this study also highlight the additional benefits of type 2 diabetes patients. Future studies are needed to investigate the benefits of continuous glucose monitoring and the effects of time on blood glucose fluctuations and range between patients with NPDR and those with type 2 diabetes.