Abstract
Aims
Time spent in the glucose range of 70–180 mg/dL (TIR) has become entrenched as a key measure of glycaemic control, which was linked to diabetes‐related outcomes in previous studies. However, there has been a recent debate about whether to instead emphasize time in the target range of 70–140 mg/dL (time in tight range, TITR). We aimed to assess the association between TITR and incident diabetic retinopathy in adults with type 2 diabetes.
Materials and Methods
This is a dynamic cohort study conducted at a tertiary hospital. 2518 adults with type 2 diabetes and without diabetic retinopathy at baseline were finally included. TITR was obtained from continuous glucose monitoring data at baseline. Cox proportional hazard regression analysis was performed to assess the relationships of TITR with the risk of incident diabetic retinopathy.
Results
During a mean follow‐up period of 5.43 years, 646 patients developed retinopathy. The multivariable‐adjusted hazard ratios (HRs) for incident retinopathy across descending TITR quartiles (Q4: >58% [reference], Q3: 38% ~ 57%, Q2: 19% ~ 37%, and Q1: <19%) were 1.00, 1.47 (95%CI 1.16, 1.87), 1.52 (95%CI 1.20, 1.93) and 1.93 (95%CI 1.53, 2.43), respectively. For per 10% decrease in TITR, the risk of diabetic retinopathy was increased by 9% (HR = 1.09, 95%CI 1.06, 1.13) after full adjustment for covariates. In the TIR >70% subgroup, the significant association between TIR, as a continuous variable and the risk of incident retinopathy disappeared, whereas TITR remained significantly associated with the outcome. Similar results were observed in the TIR >80% and TIR >90% subgroups.
Conclusions
TITR is inversely associated with the incidence of diabetic retinopathy in adults with type 2 diabetes. Among adults with well‐controlled TIR (70% or higher), TITR may provide added value regarding glucose control.
Keywords: continuous glucose monitoring, diabetic retinopathy, time in tight range, type 2 diabetes mellitus
1. INTRODUCTION
Continuous glucose monitoring (CGM) has been widely adopted in both clinical practice and clinical studies as a result of rapid technological advancement and accumulating evidence supporting its benefits. 1 , 2 , 3 , 4 The wealth of information generated by CGM has allowed for the introduction of new glycaemic metrics that can be used as a complement to glycated haemoglobin A1c (HbA1c). 5 , 6 Among these CGM‐derived metrics, time in the target range of 70–180 mg/dL (time in range, TIR) has become entrenched as a key measure of glycaemic control since the publication of the International Consensus statement in 2019. 6 , 7 It was established partly based on what could realistically be achieved as a goal for people with diabetes, considering the most current technology available at the time. However, technologies and treatments have boomed over the past few years, with the rapid development of more advanced closed‐loop automated insulin delivery systems and novel therapeutic strategies. 8 , 9 , 10 , 11 These advancements in diabetes technologies and therapies are now offering the possibility of achieving glucose levels close to euglycaemia. For example, it was shown that users of the MiniMed 780G advanced hybrid closed‐loop system can achieve an average TIR >80% when consistently using optimal system settings. 12 Therefore, new CGM metrics and targets may be warranted.
In this context, there has been a recent debate about whether to instead emphasize time spent in the glucose range of 70–140 mg/dL (time in tight range, TITR), 13 , 14 , 15 , 16 since it more closely approximates normoglycaemia. 17 TITR was introduced through the international consensus statement in 2023 and recommended as one of the CGM‐derived endpoints to be reported in clinical trials. 8 Although TITR has attracted much attention recently, there is still no evidence assessing the association of TITR with long‐term complications in people with type 2 diabetes. Therefore, based on updated data from a dynamic cohort of type 2 diabetes who were equipped with CGM at baseline, the present study aimed to investigate the association between TITR and the risk of incident diabetic retinopathy in adults with type 2 diabetes.
2. METHODS
2.1. Study design and population
This is a single‐center, observational, dynamic cohort study. Adults with type 2 diabetes admitted to the Department of Endocrinology and Metabolism at Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine were consecutively recruited between April 2005 and December 2022. Inclusion criteria for the present study were as follows: (1) type 2 diabetes; (2) age 18–80 years; (3) without retinopathy at baseline; (4) at least one follow‐up visit. Exclusion criteria included: (1) diabetic ketoacidosis or hyperglycaemic hyperosmolar state; (2) concomitant conditions that can influence glucose levels including acute infection, chemotherapy or use of corticosteroids; (3) missing data on CGM or covariates; (4) lack of follow‐up data on retinopathy or examination dates; (5) short follow‐up (<6 months). Finally, a total of 2518 individuals were included in the analyses (Figure S1). The study was approved by the Ethics Committees of Shanghai Sixth People's Hospital, and conducted in accordance with the principles of the Helsinki Declaration. Informed consent was obtained from all participants.
2.2. Baseline clinical and biological parameters
Baseline clinical data including age, sex, diabetes duration, drinking status (current or not), smoking status (current or not) and use of medication (yes or no) were collected through standardized electronic medical record forms. Meanwhile, all participants underwent a physical examination at baseline to measure blood pressure (BP), height and weight as previously described. 18 Body mass index (BMI) was then calculated as weight divided by the square of height (kg/m2). From the visits over the same period, a venous blood sample was collected for biochemical analyses at 06:00 after a 10‐h overnight fast. The details of haemoglobin A1c (HbA1c), total cholesterol (TC), triglycerides (TG), high‐density lipoprotein cholesterol (HDL‐C) and low‐density lipoprotein cholesterol (LDL‐C) measurement have been described previously. 18 The estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration equation. 19
2.3. Continuous glucose monitoring
Two retrospective CGM systems (CGMS GOLD® before September 2017 and iPro 2® thereafter, Medtronic Inc., Northridge, CA, USA) were used for subcutaneous interstitial glucose monitoring. The sensors of the CGM systems were inserted on the first day of hospital admission (due to suboptimal glycaemic control and/or the need for comprehensive examination of complications) and generated a daily record of 288 glucose values. On average, participants wore the sensor for 3 ± 1 days during their hospitalization. For calibration of the CGM systems, capillary blood glucose values were measured at least once every 12 h by using a SureStep blood glucose meter (LifeScan, Milpitas, CA, USA). CGM metrics were calculated using the CGM data during the first inpatient visit. TITR was defined as the average percentage of time spent in the target glucose range 70–140 mg/dL (3.9–7.8 mmol/L) per day (%). TIR (%), defined as the average percentage of time spent in the target glucose range 70–180 mg/dL (3.9–10.0 mmol/L) per day (%), was also calculated.
2.4. Follow‐up and outcome
The primary outcome was the incidence of diabetic retinopathy. Each participant underwent fundus photography at baseline and follow‐up by a trained ophthalmologist, who was blinded to participants' characteristics. The examination was performed using a 45°, 6.3‐megapixel nonmydriatic digital camera (CR6‐45NM; Canon, Lake Success, NY, Japan) according to a standardized protocol. Participants were promptly referred to an experienced ophthalmologist for validation and further investigation if photographs were uninterpretable or diabetic retinopathy was suspected. Diabetic retinopathy was diagnosed according to the International Classification of Diabetic Retinopathy 20 as previously described. 21 Participants were followed up as the clinicians recommended according to the guideline for the prevention and treatment of type 2 diabetes in China, 22 with a median (interquartile range) follow‐up interval of 1.53 (1.08, 2.96) years in this study. The follow‐up time was determined as from the baseline date (start date of the first inpatient visit) to the date of retinopathy diagnosis or the date of last confirmed follow‐up. The present analyses take into account outcome data of the last updating performed at March 2024.
2.5. Statistical analysis
R statistical software (version 4.2.1, https://www.r-project.org) was used for all the statistical analyses. Continuous variables and categorical variables were reported as mean ± standard deviation (SD) or n (%), respectively. Differences among groups were assessed using ANOVA tests or Jonckheere‐Terpstra tests for continuous variables, and Cochran‐Armitage trend tests for categorical variables.
Correlations among CGM parameters and HbA1c were assessed using Spearman correlation coefficients (r). Cox proportional hazards regression was performed to assess the relationship of TITR, as either categorical (based on the quartiles of TITR, with quartile 4 as the reference group) or continuous variable (per absolute 10% lower), with the presence of diabetic retinopathy (yes or no). The analyses were first carried out without adjustment, and then adjusted for sex, age, diabetes duration, family history of diabetes, current smoking status, BMI, systolic BP, TG, HDL‐C, LDL‐C and eGFR. These covariates included in the multivariable model were selected based on the univariable analysis results (p <0.1) and prior literature. 21 , 23 In sensitivity analyses, we used parametric Weibull models that incorporated interval‐censoring to account for retinopathy status being known only at study visits. 24
Additionally, we fitted a generalized additive model (GAM) with a smoothing spline term using 3 degrees of freedom to evaluate the potential nonlinear relationship of TIR and TITR, and to plot the dose–response curve. The C‐statistics of the fully adjusted Cox proportional hazards models were calculated to statistically compare the discriminatory ability between TITR and TIR (as continuous variables [per absolute 10% decrease]) in predicting the risk of retinopathy. The relationships of TIR, TITR and the risk of incident diabetic retinopathy were also assessed in subgroups stratified according to different TIR levels. A two‐tailed p value <0.05 was considered statistically significant.
3. RESULTS
A total of 2518 participants were included in the analysis. Of them, 1542 (61.2%) were male. The participants had a mean age of 56.5 ± 11.4 years, a mean BMI of 25.2 ± 3.4 kg/m2, a mean HbA1c of 8.7% ± 2.1%, and a mean TIR of 68% ± 23%. The 25th, 50th, and 75th percentiles of TITR were 19%, 38%, and 58%, respectively. The characteristics of the study population across TITR quartiles are presented in Table 1. Briefly, participants with lower TITR had a longer duration of diabetes, higher systolic BP, worse blood lipids profiles, and were more likely to receive insulin (all p for trend <0.01). Furthermore, the correlation coefficients were −0.49 for HbA1c and TITR (p <0.001), and −0.56 for HbA1c and TIR (p <0.001). We also analysed the correlations of TITR, TIR with other CGM metrics (Table S1). Notably, TITR was positively correlated with time below range (TBR) <70 mg/dL (3.9 mmol/L) (r = 0.32, p <0.001).
TABLE 1.
Baseline clinical characteristics of participants by time in tight range (TITR) quartiles.
| Total (n = 2518) | TITR Quartiles | p value for trend | ||||
|---|---|---|---|---|---|---|
| <19% (n = 617) | 19% ~ 37% (n = 635) | 38% ~ 57% (n = 622) | ≥58% (n = 644) | |||
| Sex, n (%) | 0.821 | |||||
| Male | 1542 (61.2) | 374 (60.6) | 390 (61.4) | 384 (61.7) | 394 (61.2) | |
| Female | 976 (38.8) | 243 (39.4) | 245 (38.6) | 238 (38.3) | 250 (38.8) | |
| Age, years | 56.5 ± 11.4 | 56.2 ± 11.8 | 56.9 ± 11.2 | 57.4 ± 11.2 | 55.7 ± 11.3 | 0.037 |
| Diabetes duration, years | 8.0 ± 6.4 | 8.5 ± 6.4 | 8.6 ± 6.6 | 8.4 ± 6.5 | 6.7 ± 5.7 | <0.001 |
| Systolic blood pressure, mmHg | 129 ± 16 | 130 ± 16 | 131 ± 17 | 129 ± 16 | 128 ± 15 | 0.003 |
| Diastolic blood pressure, mmHg | 80 ± 10 | 80 ± 9 | 80 ± 9 | 79 ± 10 | 80 ± 10 | 0.321 |
| BMI, kg/m2 | 25.2 ± 3.4 | 25.5 ± 3.4 | 25.1 ± 3.4 | 24.9 ± 3.5 | 25.3 ± 3.4 | 0.017 |
| Total cholesterol, mmol/L | 4.7 ± 1.1 | 4.9 ± 1.1 | 4.8 ± 1.2 | 4.7 ± 1.1 | 4.6 ± 1.0 | <0.001 |
| Triglycerides, mmol/L | 2.0 ± 1.9 | 2.4 ± 2.4 | 2.0 ± 2.2 | 1.9 ± 1.7 | 1.8 ± 1.3 | <0.001 |
| HDL‐C, mmol/L | 1.1 ± 0.3 | 1.0 ± 0.3 | 1.1 ± 0.3 | 1.1 ± 0.3 | 1.1 ± 0.3 | <0.001 |
| LDL‐C, mmol/L | 1.7 ± 1.1 | 1.6 ± 1.1 | 1.7 ± 1.2 | 1.8 ± 1.1 | 1.8 ± 1.1 | <0.001 |
| eGFR, mL/min/1.73 m2 | 96.9 ± 17.6 | 99.2 ± 18.9 | 96.9 ± 18.8 | 95.6 ± 16.6 | 96.0 ± 15.6 | <0.001 |
| HbA1c, % | 8.7 ± 2.1 | 9.6 ± 1.9 | 9.3 ± 1.9 | 8.6 ± 2.0 | 7.3 ± 1.7 | <0.001 |
| HbA1c, mmol/mol | 71.5 ± 22.4 | 81.6 ± 20.3 | 77.7 ± 21.0 | 70.6 ± 21.5 | 56.4 ± 18.1 | <0.001 |
| TIR (70–180 mg/dL), % | 68 ± 23 | 41 ± 20 | 63 ± 14 | 77 ± 11 | 91 ± 7 | <0.001 |
| Family history of diabetes, n (%) | 1365 (54.2) | 347 (56.2) | 353 (55.6) | 322 (51.8) | 343 (53.3) | 0.153 |
| Current smoker, n (%) | 735 (29.2) | 201 (32.6) | 181 (28.5) | 186 (29.9) | 167 (25.9) | 0.022 |
| Current alcohol drinker, n (%) | 490 (19.5) | 136 (22.1) | 124 (19.5) | 115 (18.5) | 115 (17.9) | 0.053 |
| Anti‐diabetic agents, n (%) | ||||||
| Oral anti‐diabetes drugs | 1841 (73.1) | 431 (69.9) | 441 (69.4) | 446 (71.7) | 523 (81.2) | <0.001 |
| Insulin | 1490 (59.2) | 438 (71.0) | 453 (71.3) | 370 (59.5) | 229 (35.6) | <0.001 |
| Anti‐hypertension agents, n (%) | 1179 (46.8) | 267 (43.3) | 313 (49.3) | 294 (47.3) | 305 (47.4) | 0.026 |
| Lipid‐lowering agents, n (%) | 1264 (50.2) | 334 (54.1) | 330 (52.0) | 308 (49.5) | 292 (45.3) | 0.001 |
Note: Data are expressed as mean ± standard deviation or n (%).
Abbreviations: BMI, body mass index; eGFR, estimated glomerular filtration rate; HbA1c, glycated haemoglobin A1c; HDL‐C, high‐density lipoprotein‐cholesterol; LDL‐C, low‐density lipoprotein‐cholesterol; TIR, time in range.
3.1. Association between TITR and DR
Over a mean follow‐up period of 5.43 ± 3.85 years, 646 participants developed retinopathy, corresponding to an incidence rate of 25.7%. TITR exhibited an inverse relationship with the presence of diabetic retinopathy. The percentages of participants with Incident diabetic retinopathy were 18.9%, 25.6%, 26.1%, and 32.3%, respectively, across descending TITR quartile groups (Q4: >58%, Q3: 38% ~ 57%, Q2: 19% ~ 37% and Q1: <19%) (p for trend<0.001) (Table 2). In the fully adjusted model (sex, age, diabetes duration, family history of diabetes, current smoking status, BMI, systolic BP, TG, HDL‐C, LDL‐C, and eGFR), the hazard ratios (HRs) for incident retinopathy across descending TITR quartiles (Q4 as the reference group) were 1.00, 1.47 (95% confidence interval [CI] 1.16, 1.87), 1.52 (95% CI 1.20, 1.93) and 1.93 (95% CI 1.53, 2.43), respectively (Table 2). For per 10% decrease in TITR, the risk of diabetic retinopathy was increased by 9% (HR = 1.09, 95% CI 1.06, 1.13) after full adjustment for covariates (Table 3). We also tested the Weibull model for a sensitivity analysis and obtained similar results (Table S2). Additionally, the significant association between TITR and retinopathy remained consistent across subgroups, including males and females, those ≥60 years and <60 years, and insulin users and non‐users, with no significant heterogeneity between these groups (all p for interaction >0.05) (Table S3).
TABLE 2.
Hazard ratios (HR) for incident diabetic retinopathy according to time in tight range (TITR) quartiles.
| TITR quartiles | ||||
|---|---|---|---|---|
| Q1 (<19%) | Q2 (19% ~ 37%) | Q3 (38% ~ 57%) | Q4 (≥58%) | |
| No. of participants | 617 | 635 | 622 | 644 |
| No. (%) with outcome | 199 (32.3) | 166 (26.1) | 159 (25.6) | 122 (18.9) |
| Unadjusted HRs (95%CIs) | 2.16 (1.73, 2.71) | 1.66 (1.31, 2.09) | 1.56 (1.23, 1.97) | 1.00 |
| Adjusted HRs (95%CIs) a | 1.93 (1.53, 2.43) | 1.52 (1.20, 1.93) | 1.47 (1.16, 1.87) | 1.00 |
Adjusted for sex, age, diabetes duration, family history of diabetes, current smoking status, body mass index, systolic blood pressure, triglycerides, high‐density lipoprotein cholesterol, low‐density lipoprotein cholesterol and estimated glomerular filtration rate.
TABLE 3.
Comparison of overall prediction performance between time in range (TIR) and time in tight range (TITR) in the risk of incident diabetic retinopathy.
| Per 10% lower TIR | Per 10% lower TITR | |
|---|---|---|
| Unadjusted HRs (95%CIs) | 1.12 (1.08, 1.15) | 1.11 (1.08, 1.15) |
| Adjusted HRs (95%CIs) a | 1.09 (1.06, 1.12) | 1.09 (1.06, 1.13) |
| C‐statistic (95% CI) a | 0.565 (0.539, 0.591) | 0.563 (0.537, 0.588) |
| p for change in C‐statistic | Ref. | 0.75 |
Note: TIR and TITR were included in the models as continuous variables (per 10% decrease).
Abbreviation: HR, hazard ratio.
Adjusted for sex, age, diabetes duration, family history of diabetes, current smoking status, body mass index, systolic blood pressure, triglycerides, high‐density lipoprotein cholesterol, low‐density lipoprotein cholesterol and estimated glomerular filtration rate.
3.2. Comparison between TITR and TIR
TITR was highly correlated with TIR (r = 0.85, p <0.001). However, the relationship between TIR and TITR was nonlinear, with a higher ratio of TITR/TIR observed as TIR increased (Figure S2, Table S4). Meanwhile, an increasingly wide range of TITR values for a given TIR level was also observed, as TIR increased (Table S4).
Overall, the C‐statistics of the fully adjusted models incorporating TITR and TIR were 0.563 and 0.565, respectively (p = 0.75) (Table 3), indicating that both metrics have comparable predictive value for diabetic retinopathy. Subgroup analyses were performed across varying TIR levels. As shown in Figure 1, in the TIR >40%, TIR >50% and TIR >60% subgroups, both TIR and TITR were significantly associated with the risk of incident diabetic retinopathy in the fully adjusted models. However, when the cutoff points for TIR increased to 70% or higher, the significant association between TIR and the risk of incident diabetic retinopathy disappeared, whereas TITR remained significantly associated with the outcome. Specifically, in the TIR >70%, TIR >80% and TIR >90% subgroups, with per 10% decrease in TITR, the HRs (95% CIs) of incident diabetic retinopathy were 1.09 (1.02, 1.15), 1.10 (1.02, 1.18) and 1.15 (1.03, 1.30), respectively.
FIGURE 1.
Hazard ratios for incident diabetic retinopathy according to time in tight range (TITR) and time in range (TIR) in different TIR subgroups. †Adjusted for sex, age, diabetes duration, family history of diabetes, current smoking status, body mass index, systolic blood pressure, triglycerides, high‐density lipoprotein cholesterol, low‐density lipoprotein cholesterol and estimated glomerular filtration rate. TIR and TITR were included in the models as continuous variables (per 10% decrease). HR, hazard ratio.
4. DISCUSSION
In this large CGM‐based cohort study, we provide evidence that lower TITR is significantly associated with an increased risk of incident diabetic retinopathy in adults with type 2 diabetes. Moreover, we found that in the TIR >70%, TIR >80% and TIR >90% subgroups, the significant association between TIR and diabetic retinopathy disappeared, whereas TITR continued to exhibit a significant correlation with diabetic retinopathy.
Given that TITR is a relatively new metric, currently evidence linking TITR to diabetes‐related outcomes is very limited. In a longitudinal study of 163 adults with type 1 diabetes, up to 7 years of CGM data were analysed, and TITR was found to be associated with an increased risk of incident diabetic retinopathy. 25 In another cross‐sectional study involving 808 individuals with type 1 diabetes, TITR was shown to be inversely associated with the presence of both microvascular complications and cerebrovascular disease. 26 However, there is no evidence linking TITR to long‐term outcomes among people with type 2 diabetes. To our knowledge, this study is the first to demonstrate a significant association between TITR and microvascular complications in adults with type 2 diabetes.
It is important to understand the relationship between TITR and TIR. Previous studies have shown a strong correlation between TITR and TIR as expected (r ≥0.94). 25 , 27 , 28 Additionally, TITR was found to be consistently 20%–25% lower than TIR on average across different diabetic cohorts and with different insulin delivery methods. 15 Therefore, it is not surprising that overall, TITR and TIR provide comparable predictive value for diabetic retinopathy risk in our study, which is consistent with previously reported findings. 25 However, it should be noted that a non‐linear relationship between TIR and TITR has been reported in previous studies, 27 , 29 with a higher ratio of TITR/TIR observed as TIR increased, 27 which was also confirmed by our study. More importantly, a previous study using real‐world data from over 20 000 CGM users demonstrated from a mathematical perspective that TITR appears to be better when normal or near‐normal glycaemia is achieved, as TIR is insensitive to glycaemic changes in this situation. 16 Based on these findings, we performed subgroups analyses and found that in individuals with TIR above 70% or higher, the significant association between TIR and diabetic retinopathy disappeared, whereas TITR remained significantly correlated with outcome. From an epidemiological standpoint, our results suggest that TITR may be a better predictor of complications and hence more valuable than TIR when TIR approaches 70% or higher. Conversely, it is reasonable to speculate that when TITR is low, TIR may provide better discrimination of higher complication risk groups. Therefore, TITR and TITR should be considered as complementary measures in clinical practice. However, it should be noted that the average CGM duration was 3 days in the current study, therefore future studies with longer days of CGM data are needed to validate these findings.
Consistently, we previously noted that glucose levels above 140 or 150 mg/dL have a meaningful impact on the risk of complications. In a cross‐sectional analysis of 2893 individuals with type 2 diabetes, we found that TARs/TIRs with the upper limit ranging from 140–150 to 200 mg/dL (7.8–8.3 to 11.1 mmol/L) were all significantly associated with abnormal carotid intima‐media thickness and diabetic retinopathy. 30 Therefore, achieving a tighter glucose target using metrics such as TITR may logically help further reduce the risk of complications. However, there are concerns and potential challenges when using TITR in clinical practice. 14 A recent survey examining the experience of CGM metrics among paediatric CGM user and their parents showed that participants expressed a range of concerns about TITR, including increased stress, risk of hypoglycaemia, self‐criticism and family conflict. And participants in this study also requested more clinical support and a clear scientific rationale for changes to glycaemic target ranges. 31 Moreover, in the current study, a significant positive correlation was observed between TITR and TBR <70 mg/dL (3.9 mmol/L). Thus, as a tighter glucose target, the implementation of TITR, if set too high, might be accompanied by an increase in hypoglycaemia risk, which is a major concern also highlighted by Hamidi et al. 14 Therefore, there is a need to develop education programs and clinical support if new glucose targets are introduced. Besides, the name TITR should probably be modified to time in normoglycaemia or time in the non‐diabetes range, which is less threatening.
The main strengths of this study include cohort design, long follow‐up period and the relatively large sample size. However, there are several limitations to this study. First, most of the participants underwent CGM for 3 days, which might not optimally reflect the long‐term glycaemic patterns in real life, as discussed in our previous study. 18 Therefore, the current results should be interpreted with caution, and longer days of CGM data, along with updated mean TITR, are needed to validate the results. Nevertheless, when pooling data over a large cohort, as in the present study, the measurement errors may be reduced, such that some CGM metrics may be reliably assessed from CGM data with a much shorter duration. 32 Moreover, of note, a previous study showed that for TIR, a strong correlation to the full 3 months of data was observed after 3 days of sampling, with the R 2 increasing from 0.77 to 0.88 as the sampling duration extended from 3 to 14 days, as shown in Data S1. 33 Second, the index date of the outcome was determined by the date of the fundus examination. Consequently, it was not possible to accurately capture the exact date of incident diabetic retinopathy. However, the median (interquartile range) follow‐up interval was 1.53 (1.08, 2.96) years, which conformed to the screening frequency for retinopathy recommended by the guideline in China. 22 Third, due to the lack of detailed data on diabetic retinopathy grading, we were unable to analyse the relationship between TITR and varying degrees of diabetic retinopathy. Finally, the participants enrolled were all hospital‐based Chinese with type 2 diabetes. It remains to be explored whether the present findings could be generalized to other diabetic populations.
In conclusion, lower TITR as measured by CGM was associated with an increased risk of incident diabetic retinopathy in adults with type 2 diabetes. Moreover, we observed that among adults with well‐controlled TIR (70% or higher), TITR appears more informative and may provide added value in assessing the risk of complications, which need to be verified in future studies. Therefore, our results provide evidence that TITR is correlated with long‐term diabetic complications in people with type 2 diabetes. Future research is warranted to determine whether a TITR goal could be safely achieved without an increase in hypoglycaemia.
AUTHOR CONTRIBUTIONS
JZ designed the study. JY, JN, MW, WL, WZ and YW collected the data and reviewing it critically for important intellectual content. YW and JN cleaned the data. YW and JL performed statistical analysis and wrote the draft of the manuscript. JG provided advices for the data analyses. YB and JZ reviewed and edited the manuscript. All authors read and approved the final manuscript. JZ is the guarantor of this work.
FUNDING INFORMATION
This work was funded by the Program of Shanghai Academic Research Leader (22XD1402300), the Shanghai Oriental Talent Program (Youth Project) (No. NA), the Shanghai Research Center for Endocrine and Metabolic Diseases (2022ZZ01002), the National Key Clinical Specialty (Z155080000004) and the Shanghai Key Discipline of Public Health Grants Award (GWVI‐11.1‐20 and GWVI‐11.1‐22).
CONFLICT OF INTEREST STATEMENT
All the authors have no conflict of interest to declare.
PEER REVIEW
The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/dom.16143.
ETHICS STATEMENT
The study and the analysis plan were approved by the Ethics Committees of Shanghai Sixth People's Hospital. We have obtained informed consent from all participants.
Supporting information
Data S1. Supporting information.
ACKNOWLEDGEMENTS
The authors appreciate Robert A. Vigersky (Medtronic Diabetes, Northridge, CA) for dedicating his time and expertise to the completion of this study. The authors also appreciate all students, research staff and patients who participated in this study.
Wang Y, Lu J, Yu J, et al. Association between time in tight range and incident diabetic retinopathy in adults with type 2 diabetes. Diabetes Obes Metab. 2025;27(3):1415‐1422. doi: 10.1111/dom.16143
Yaxin Wang and Jingyi Lu contributed equally to this study.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
REFERENCES
- 1. Beck RW, Riddlesworth T, Ruedy K, et al. Effect of continuous glucose monitoring on glycemic control in adults with type 1 diabetes using insulin injections: the DIAMOND randomized clinical trial. JAMA. 2017;317(4):371‐378. [DOI] [PubMed] [Google Scholar]
- 2. Lind M, Polonsky W, Hirsch IB, et al. Continuous glucose monitoring vs conventional therapy for glycemic control in adults with type 1 diabetes treated with multiple daily insulin injections: the GOLD randomized clinical trial. JAMA. 2017;317(4):379‐387. [DOI] [PubMed] [Google Scholar]
- 3. Ajjan RA, Battelino T, Cos X, et al. Continuous glucose monitoring for the routine care of type 2 diabetes mellitus. Nat Rev Endocrinol. 2024;20(7):426‐440. [DOI] [PubMed] [Google Scholar]
- 4. 7. Diabetes technology: standards of Care in Diabetes‐2024. Diabetes Care. 2024;47(Suppl 1):S126‐S144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Battelino T, Alexander CM, Amiel SA, et al. Continuous glucose monitoring and metrics for clinical trials: an international consensus statement. Lancet Diabetes Endocrinol. 2023;11(1):42‐57. [DOI] [PubMed] [Google Scholar]
- 6. American Diabetes Association Professional Practice Committee . 6. Glycemic goals and hypoglycemia: standards of Care in Diabetes‐2024. Diabetes Care. 2024;47(Suppl 1):S111‐S125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Battelino T, Danne T, Bergenstal RM, et al. Clinical targets for continuous glucose monitoring data interpretation: recommendations from the international consensus on time in range. Diabetes Care. 2019;42(8):1593‐1603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Phillip M, Nimri R, Bergenstal RM, et al. Consensus recommendations for the use of automated insulin delivery technologies in clinical practice. Endocr Rev. 2023;44(2):254‐280. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Daly AB, Boughton CK, Nwokolo M, et al. Fully automated closed‐loop insulin delivery in adults with type 2 diabetes: an open‐label, single‐center, randomized crossover trial. Nat Med. 2023;29(1):203‐208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Lee TTM, Collett C, Bergford S, et al. Automated insulin delivery in women with pregnancy complicated by type 1 diabetes. N Engl J Med. 2023;389(17):1566‐1578. [DOI] [PubMed] [Google Scholar]
- 11. Ceriello A, Prattichizzo F, Phillip M, Hirsch IB, Mathieu C, Battelino T. Glycaemic management in diabetes: old and new approaches. Lancet Diabetes Endocrinol. 2022;10(1):75‐84. [DOI] [PubMed] [Google Scholar]
- 12. Arrieta A, Battelino T, Scaramuzza AE, et al. Comparison of MiniMed 780G system performance in users aged younger and older than 15 years: evidence from 12 870 real‐world users. Diabetes Obes Metab. 2022;24(7):1370‐1379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Akturk HK, Battelino T, Castañeda J, Arrieta A, van den Heuvel T, Cohen O. Future of time‐in‐range goals in the era of advanced hybrid closed‐loop automated insulin delivery systems. Diabetes Technol Ther. 2024;26(S3):102‐106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Hamidi V, Pettus JH. Time in tight range for patients with type 1 diabetes: the time is now, or is it too soon? Diabetes Care. 2024;47(5):782‐784. [DOI] [PubMed] [Google Scholar]
- 15. Beck RW. Is it time to replace time‐in‐range with time‐in‐tight‐range? Maybe not. Diabetes Technol Ther. 2024;26(3):147‐150. [DOI] [PubMed] [Google Scholar]
- 16. Dunn TC, Ajjan RA, Bergenstal RM, Xu Y. Is it time to move beyond TIR to TITR? Real‐world data from over 20,000 users of continuous glucose monitoring in patients with type 1 and type 2 diabetes. Diabetes Technol Ther. 2024;26(3):203‐210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Shah VN, DuBose SN, Li Z, et al. Continuous glucose monitoring profiles in healthy nondiabetic participants: a multicenter prospective study. J Clin Endocrinol Metab. 2019;104(10):4356‐4364. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Lu J, Wang C, Shen Y, et al. Time in range in relation to all‐cause and cardiovascular mortality in patients with type 2 diabetes: a prospective cohort study. Diabetes Care. 2021;44(2):549‐555. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604‐612. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Wilkinson CP, Ferris FL 3rd, Klein RE, et al. Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales. Ophthalmology. 2003;110(9):1677‐1682. [DOI] [PubMed] [Google Scholar]
- 21. Lu J, Ma X, Zhou J, et al. Association of Time in range, as assessed by continuous glucose monitoring, with diabetic retinopathy in type 2 diabetes. Diabetes Care. 2018;41(11):2370‐2376. [DOI] [PubMed] [Google Scholar]
- 22. Jia W, Weng J, Zhu D, et al. Standards of medical care for type 2 diabetes in China 2019. Diabetes Metab Res Rev. 2019;35(6):e3158. [DOI] [PubMed] [Google Scholar]
- 23. Wang Y, Lu J, Shen Y, et al. Comparison of glucose time in range and area under curve in range in relation to risk of diabetic retinopathy in type 2 diabetes patients. J Diabetes Investig. 2022;13(9):1543‐1550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Beiki O, Frumento P, Bottai M, Manouchehrinia A, Hillert J. Changes in the risk of reaching multiple sclerosis disability milestones in recent decades: a nationwide population‐based cohort study in Sweden. JAMA Neurol. 2019;76(6):665‐671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Shah VN, Kanapka LG, Akturk HK, et al. Time in range is associated with incident diabetic retinopathy in adults with type 1 diabetes: a longitudinal study. Diabetes Technol Ther. 2024;26(4):246‐251. [DOI] [PubMed] [Google Scholar]
- 26. De Meulemeester J, Charleer S, Visser MM, De Block C, Mathieu C, Gillard P. The association of chronic complications with time in tight range and time in range in people with type 1 diabetes: a retrospective cross‐sectional real‐world study. Diabetologia. 2024;67:1527‐1535. [DOI] [PubMed] [Google Scholar]
- 27. Beck RW, Raghinaru D, Calhoun P, Bergenstal RM. A comparison of continuous glucose monitoring‐measured time‐in‐range 70‐180 mg/dL versus time‐in‐tight‐range 70‐140 mg/dL. Diabetes Technol Ther. 2024;26(3):151‐155. [DOI] [PubMed] [Google Scholar]
- 28. Passanisi S, Piona C, Salzano G, et al. Aiming for the best glycemic control beyond time in range: time in tight range as a new continuous glucose monitoring metric in children and adolescents with type 1 diabetes using different treatment modalities. Diabetes Technol Ther. 2024;26(3):161‐166. [DOI] [PubMed] [Google Scholar]
- 29. Castañeda J, Arrieta A, van den Heuvel T, Battelino T, Cohen O. Time in tight glucose range in type 1 diabetes: predictive factors and achievable targets in real‐world users of the MiniMed 780G system. Diabetes Care. 2024;47(5):790‐797. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Lu J, Home PD, Zhou J. Comparison of multiple cut points for time in range in relation to risk of abnormal carotid intima‐media thickness and diabetic retinopathy. Diabetes Care. 2020;43(8):e99‐e101. [DOI] [PubMed] [Google Scholar]
- 31. Tanenbaum ML, Pang E, Tam R, et al. ‘We're taught green is good’: perspectives on time in range and time in tight range from youth with type 1 diabetes, and parents of youth with type 1 diabetes. Diabet Med. 2024;41:e15423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Foreman YD, Brouwers M, van der Kallen CJH, et al. Glucose variability assessed with continuous glucose monitoring: reliability, reference values, and correlations with established glycemic indices‐the Maastricht study. Diabetes Technol Ther. 2020;22(5):395‐403. [DOI] [PubMed] [Google Scholar]
- 33. Riddlesworth TD, Beck RW, Gal RL, et al. Optimal sampling duration for continuous glucose monitoring to determine long‐term glycemic control. Diabetes Technol Ther. 2018;20(4):314‐316. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1. Supporting information.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.