Association of Time in Tight Range with Microvascular Complications in Japanese Patients with Type 2 Diabetes Mellitus: A Multicenter Cross-Sectional Study

Abstract

Background:

To determine the correlation between time in tight range (TITR; 70–140 mg/dL) and prevalence of microvascular complications in patients with type 2 diabetes mellitus (T2DM).

Methods:

Data of 999 patients with T2DM and negative history of cardiovascular disease were analyzed. TITR was assessed using data from a continuous glucose monitoring (CGM) system. Participants were stratified into quartiles based on TITR (Q1: ≤42.6%, Q2: >42.6 to ≤61.2%, Q3: >61.2 to ≤73.4%, Q4: >73.4%). The correlation of TITR/microvascular complications was assessed using multivariate logistic regression analysis after adjustment for potential confounders.

Results:

The mean TITR was 56.8 ± 22.4%, and 51.2% of participants had at least one microvascular complication. The adjusted odds ratios for any microvascular complication across increasing TITR quartiles were 1.00 (Q1 as the reference group), 0.39 (Q2; 95% confidence interval [CI]: 0.25–0.62), 0.45 (Q3; 95% CI: 0.28–0.70), and 0.30 (Q4; 95% CI: 0.19–0.47). This indicated that the prevalence of diabetic microvasculopathies was lower in higher TITR quartiles. Similar inverse trends were observed for retinopathy, nephropathy, and peripheral neuropathy. Each 10% increase in TITR was associated with a reduced risk of each type of diabetic microvasculopathy. Receiver operating characteristic curve analysis identified 54.3% as the optimal TITR cutoff value for the identification of microvascular complications.

Conclusions:

Higher TITR was significantly associated with lower prevalence of microvascular complications in patients with T2DM. CGM-derived TITR is a potentially useful clinical metric for optimizing glycemic management and reducing the risk of microvascular complications.

Trial Registration Number:

UMIN000032325.

Keywords

type 2 diabetes microvascular complications continuous glucose monitoring time in tight range

Introduction

The risk of serious complications in diabetes is closely associated with glycemic control. Elevated glycated hemoglobin (HbA1c) levels are strongly associated with the onset and progression of various microvascular complications, including retinopathy, nephropathy, and peripheral neuropathy.¹ However, since HbA1c reflects the average blood glucose level over the preceding 2–3 months, this parameter does not adequately capture short-term glycemic fluctuations or the frequencies of hypoglycemia and hyperglycemia. Furthermore, HbA1c can also be affected by certain conditions apart from diabetes, such as anemia and hemoglobinopathies.² Consequently, the time in range (TIR), which is a metric derived from continuous glucose monitoring (CGM), has garnered attention as a complementary indicator of HbA1c, with the 2019 International Consensus Guidelines recommending it as a clinical target.³ Several studies have reported the association of TIR with microvascular and cardiovascular complications in patients with type 2 diabetes mellitus (T2DM)^4–6 Our group also reported previously that TIR is associated with the severity of diabetic retinopathy and albuminuria.⁷ However, the sensitivity of TIR seems to decrease when glycemic control approaches normal levels,⁸ which limits its utility in a more stringent glycemic management, including in patients with well-controlled diabetes. Recent advances in T2DM treatment and approaches have led to recommendations for tighter glycemic control, even in patients with favorable HbA1c levels.^9,10 In this context, time in tight range (TITR), which is defined as time spent in a narrow glucose range of 70–140 mg/dL, presents a more sensitive indicator of near-normoglycemia.¹¹ Accordingly, TITR may allow a more precise evaluation of the risk of microvascular complications as it allows a better capture of glycemic excursions closer to the normoglycemic range, compared with TIR.¹²

Studies on the association of TITR with diabetic complications have so far been primarily limited to patients with type 1 diabetes mellitus (T1DM),^13,14 with little, if any, involving patients with T2DM. A recent study found that a lower TITR was independently associated with incident diabetic retinopathy in patients with T2DM.¹⁵ Notably, TITR remained significantly associated with the risk of retinopathy even among patients with well-controlled TIR (>70%), suggesting its utility in more granular risk stratification.

Apart from retinopathy, a few studies have examined comprehensively the association of TITR with a broader spectrum of microvascular complications, including nephropathy and peripheral neuropathy in patients with T2DM. However, the optimal TITR cutoff value for the identification of these complications remains unclear.

The present study was designed to determine the cross-sectional association between TITR and the prevalence of microvascular complications in order to determine the clinical utility of TITR in patients with T2DM.

Materials and Methods

Study design and participants

The present study was a subanalysis of a multicenter, prospective observational cohort study on the association between CGM-derived glycemic indices and the incidence of cardiovascular events in Japanese patients with T2DM negative for history of cardiovascular disease (CVD).¹⁶ The above prospective study included 1000 outpatients from 34 clinical sites across Japan, with a planned follow-up period of 10 years. In the present study, we performed a cross-sectional analysis using baseline data to evaluate the association between CGM-derived TITR and the prevalence of microvascular complications.

The inclusion and exclusion criteria were described in detail previously.¹⁶ Briefly, eligible participants were patients with T2DM aged 30–80 years who had no history of CVD and whose diabetes treatment regimen had remained unchanged for ≥6 months prior to the provision of written informed consent. Among the 1000 participants who met the eligibility criteria between May 2018 and March 2019, 1 withdrew consent.

The study was conducted in accordance with the principles of the Declaration of Helsinki, and the study protocol was approved by the ethics committees of all participating institutions. All participants provided written informed consent.

Anthropometric and biochemical measurements

Data regarding diabetes duration, smoking status, comorbidities, and medication use were collected from the medical records and relevant questionnaires. Anthropometric and clinical assessments, including measurements of body height, weight, and blood pressure, were performed using standardized protocols. Blood samples were collected after an overnight fast. The body mass index (BMI) was calculated as weight (kg) divided by height in meters squared (m²). Laboratory measurements of HbA1c, glucose, total cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides, creatinine, and uric acid were performed using standard methods. The estimated glomerular filtration rate (eGFR; mL/min/1.73 m²) was calculated using the equation recommended by the Japanese Society of Nephrology.¹⁷ The urine albumin-to-creatinine ratio (UACR) was measured using spot urine samples.

Assessment of microvascular complications

Certified ophthalmologists assessed the presence and severity of diabetic retinopathy, with classification into four categories based on the data available in the medical records as either no diabetic retinopathy, simple diabetic retinopathy, preproliferative diabetic retinopathy, or proliferative diabetic retinopathy. Diabetic nephropathy was defined based on UACR levels as follows: normoalbuminuria (<30 mg/gCr), microalbuminuria (30 to <300 mg/gCr), and macroalbuminuria (≥300 mg/gCr). Diabetic peripheral neuropathy was diagnosed in our subjects when two or more of the following three criteria were met: (1) bilateral presence of numbness, pain, or dysesthesia in the toes or soles; (2) bilateral reduced or absent Achilles tendon reflexes; and (3) bilateral reduced vibration sensation at the medial malleolus or evident abnormalities in nerve conduction studies.

Continuous glucose monitoring

CGM was performed using the FreeStyle Libre Pro system (Abbott Japan, Tokyo, Japan) over a nominal period of 14 days. Based on the inherent flaw of reduced sensor accuracy during the initial 24 h and the final 4 days of the monitoring period,¹⁸ only subjected glucose data from days 3 to 10 (measured at 15-min intervals) were subjected to analysis. The following CGM-derived metrics were calculated: glucose coefficient of variation, standard deviation, TIR (70–180 mg/dL), TITR (70–140 mg/dL), time above the range (TAR, >180 mg/dL), and time below the range (TBR, <70 mg/dL).

Statistical analysis

Continuous variables were presented as either mean ± standard deviation (SD) values or median values with interquartile ranges for normally distributed data and data that showed a skewed distribution pattern, respectively. Categorical variables were expressed as counts and percentages. First, the baseline characteristics of all participants (n = 999) were compared based on the presence or absence of microvascular complications. Subsequently, participants were stratified into quartiles based on TITR values (25th, 50th, and 75th percentiles), followed by among-group comparisons of the prevalence of complications. Furthermore, we compared the prevalence of complications between patients who achieved a TITR target value of >50% and those who did not. The selection of 50% TITR target was based on the recent study of Akturk et al.¹⁹ The Student’s t-test and the Mann–Whitney U test were used for comparisons of variables with normal distribution and those with skewed distribution, respectively. The χ² test was used for comparisons of categorical variables. For trends across more than three groups, the Jonckheere–Terpstra and Cochran–Armitage tests were used for continuous and categorical variables, respectively. Logistic regression analyses were performed to estimate the odds ratios (ORs) for microvascular complications according to TITR quartile (Q1 as the reference) and per 10% increase in TITR. Multivariate models were adjusted for sex, age, diabetes duration, BMI, systolic blood pressure (SBP), total cholesterol, log-transformed triglycerides, HDL cholesterol, eGFR, uric acid, log-transformed UACR, smoking status, alcohol consumption, and use of medications, including insulin, ACE inhibitors, angiotensin II receptor blockers, and statins (plus HbA1c). Receiver operating characteristic (ROC) curve analysis was performed to establish the optimal TITR cutoff value for the identification of microvascular complications. Furthermore, the area under the curve (AUC), with 95% confidence intervals (95% CI), was calculated. To assess the potential effect of baseline heterogeneity in treatment modalities and diabetes duration, we conducted interaction analyses by including cross-product terms (TITR quartiles × insulin therapy status and TITR quartiles × diabetes duration category) in the multivariable logistic regression models. The estimated ORs and 95% CIs were computed, and interaction p-values were obtained from likelihood ratio tests that compared models with and without the interaction terms. All statistical analyses were conducted using the SAS software (version 9.4; SAS Institute, Cary, NC). Statistical significance was set at P < 0.05.

Results

Clinical characteristics of participants

Supplementary Table S1 summarizes the baseline clinical characteristics of the 999 participants. The mean age was 64.6 ± 9.6 years, 60.9% were males, and the mean BMI was 24.6 ± 3.9 kg/m². The mean HbA1c was 7.1 ± 0.8%. The average duration of diabetes was 12.9 ± 8.5 years. Among the participants, 89.5% and 15.8% were being treated with glucose-lowering agents and insulin, respectively. Regarding the CGM-derived metrics, the mean TITR, TIR, average glucose level, TAR, and TBR were 56.8 ± 22.4%, 78.9 ± 18.6%, 140.5 ± 32.3 mg/dL, 19.0 ± 19.2%, and 2.16 ± 4.71%, respectively. Overall, 51.2% of the participants (n = 511) were diagnosed with at least one microvascular complication; specifically, 22.2% (n = 222) had diabetic retinopathy (13.3% simple retinopathy, 5.0% preproliferative, and 3.9% proliferative), 27.0% (n = 270) had diabetic nephropathy, and 28.6% (n = 286) had diabetic peripheral neuropathy.

Patients’ characteristics according to absence/presence of microvasculopathies

Supplementary Table S1 shows the differences in clinical characteristics between participants with and without microvasculopathies. Compared with participants free of such complications, those with microvasculopathies were older, had a longer duration of diabetes, and were more likely to be on glucose-lowering medications and renin–angiotensin–aldosterone system inhibitors (all P < 0.001). Furthermore, they had higher HbA1c levels as well as significantly lower TITR and TIR values (all P < 0.001), compared with complication-free diabetics.

TITR and microvascular complications

Participants were stratified into quartiles based on their TITR values (Q1–Q4; Table 1). Participants of the lowest quartile had a longer duration of diabetes, less favorable lipid profiles, and higher use of GLP-1 receptor agonists and insulin (all Ptrend < 0.001). Supplementary Table S2 shows the distributions of TITR and TIR across HbA1c levels, with both metrics being significantly lower in patients with higher HbA1c levels (P < 0.001).

Table 1.

Patient Characteristics According to Time in Tight Range Quartiles

Variables	TITR quartiles
Variables	Q1 ≤ 42.6%	Q2 > 42.6% to ≤61.2%	Q3 > 61.2% to ≤73.4%	Q4 > 73.4%	P value for trend
n	250	250	247	252
Age, years	63.9 ± 10.5	66.5 ± 8.7	64.6 ± 9.7	63.2 ± 9.3	0.041
Gender, males/females	159/91	147/103	154/93	148/104	0.420^*
Duration of diabetes, years	13.9 ± 8.1	15.0 ± 9.1	12.2 ± 8.5	10.3 ± 7.6	<0.001
Body mass index, kg/m²	25.1 ± 4.1	24.3 ± 3.6	24.3 ± 3.6	24.7 ± 4.0	0.291
Systolic blood pressure, mmHg	130.1 ± 15.4	131.8 ± 13.9	130.8 ± 14.9	132.1 ± 15.1	0.103
Diastolic blood pressure, mmHg	75.1 ± 11.3	74.4 ± 11.1	75.3 ± 10.1	77.3 ± 11.4	0.003
HbA1c, %	7.8 ± 0.9	7.2 ± 0.6	6.8 ± 0.4	6.5 ± 0.4	<0.001
Lipid profile
Total cholesterol, mg/dL	187.4 ± 34.4	185.2 ± 30.0	184.2 ± 32.2	186.5 ± 29.7	0.680
LDL cholesterol, mg/dL	104.2 ± 28.9	102.6 ± 25.8	103.3 ± 25.0	102.4 ± 26.2	0.558
HDL cholesterol, mg/dL	58.4 ± 16.6	59.4 ± 13.8	60.2 ± 14.7	63.4 ± 17.0	<0.001
Triglycerides, mg/dL	106 [79.0, 150.0]	102 [72.0, 154.0]	98 [70.0, 134.0]	90 [65.0, 126.5]	<0.001
Uric acid, mg/dL	5.1 ± 1.2	5.1 ± 1.3	5.2 ± 1.2	5.3 ± 1.2	0.197
eGFR, mL/min/1.73 m²	74.6 ± 25.0	72.1 ± 19.8	74.6 ± 19.4	72.2 ± 17.6	0.834
u-Alb/Cr, mg/g・Cre	174.7 ± 594.3	82.2 ± 280.5	77.6 ± 288.9	53.1 ± 182.7	<0.001
Use of oral glucose-lowering agents, n (%)	232 (92.8)	232 (92.8)	217 (87.9)	213 (84.5)	<0.001^*
Metformin	129 (51.6)	148 (59.2)	140 (56.7)	126 (50.0)	0.597^*
Sulfonylurea	42 (16.8)	46 (18.4)	26 (10.5)	13 (5.2)	<0.001^*
Glinide	24 (9.6)	21 (8.4)	14 (5.7)	9 (3.6)	0.003^*
Dipeptidyl peptidase-4 inhibitors	143 (57.2)	155 (62.0)	145 (58.7)	134 (53.2)	0.268^*
Sodium-glucose cotransporter-2 inhibitors	71 (28.4)	62 (24.8)	54 (21.9)	44 (17.5)	0.003^*
Thiazolidinediones	38 (15.2)	44 (17.6)	38 (15.4)	23 (9.1)	0.038^*
α-glucosidase inhibitor	43 (17.2)	41 (16.4)	31 (12.6)	57 (22.6)	0.239^*
Glucagon-like peptide-1 antagonists	31 (12.4)	20 (8.0)	14 (5.7)	9 (3.6)	<0.001^*
Insulin	69 (27.6)	54 (21.6)	25 (10.1)	10 (4.0)	<0.001^*
Use of antihypertensive drugs, n (%)	116 (46.4)	133 (53.2)	112 (45.3)	122 (48.4)	0.829^*
ACE inhibitors	8 (3.2)	10 (4.0)	6 (2.4)	4 (1.6)	0.169^*
Angiotensin II receptor blockers	92 (36.8)	104 (41.6)	91 (36.8)	103 (40.9)	0.586^*
Calcium channel blockers	64 (25.6)	75 (30.0)	64 (25.9)	70 (27.8)	0.845^*
Use of lipid-lowering agents, n (%)	135 (54.0)	148 (59.7)	151 (61.1)	161 (63.9)	0.025^*
Statins	123 (49.2)	127 (51.2)	126 (51.0)	132 (52.4)	0.842^*
Ezetimibe	13 (5.2)	26 (10.5)	34 (13.8)	34 (13.5)	0.001^*
Fibrates	9 (3.6)	12 (4.8)	9 (3.6)	11 (4.4)	0.842^*
Use of antithrombotic agents, n (%)	20 (8.0)	10 (4.0)	18 (7.3)	16 (6.3)	0.809
Antiplatelet agents	17 (6.8)	6 (2.4)	15 (6.1)	12 (4.8)	0.690^*
Anticoagulants	3 (1.2)	4 (1.6)	3 (1.2)	5 (2.0)	0.565^*
FLP-CGM-derived metrics
TITR, %	25.0 ± 12.9	53.0 ± 5.3	67.4 ± 3.5	81.8 ± 6.1	<0.001
TIR, %	53.7 ± 18.0	79.2 ± 7.6	88.2 ± 5.2	94.4 ± 3.9	<0.001
Average, mg/dL	182.5 ± 32.0	142.3 ± 11.3	125.7 ± 9.9	111.4 ± 8.9	<0.001
SD, mg/dL	46.6 ± 12.1	39.8 ± 8.7	34.3 ± 6.3	26.3 ± 5.2	<0.001
CV, %	25.8 ± 6.3	28.1 ± 6.1	27.4 ± 5.2	23.5 ± 4.2	<0.001
TAR, %	45.3 ± 18.4	18.7 ± 7.2	9.1 ± 4.6	2.9 ± 2.5	<0.001
TBR, %	1.00 ± 3.84	2.12 ± 5.73	2.77 ± 5.05	2.75 ± 3.72	<0.001

Data are mean ± SD, median [25%tile, 75%tile] or n (%).

P values indicate the trend for TITR Quartiles by Jonckheere–Terpstra trend test.

P values indicate the trend for TITR Quartiles by Cochran–Armitage trend test.

CV, coefficient of variation; HbA1c, hemoglobin A1c; SD, standard deviation; TAR, time above range; TBR, time below range; TIR, time in range; TITR, time in tight range.

Table 2 and Figure 1 show the prevalence of microvascular complications according to the TITR quartiles. TITR was inversely associated with the prevalence of diabetic microvasculopathies. For the lowest (Q1, TITR ≤42.6%) and highest (Q4, TITR >73.4%) quartiles, 70.4% (176/250) and 35.3% (89/252) of the participants had at least one complication, respectively. Relative to Q1, the adjusted ORs for any complication were 0.39 (95% confidence interval [95%CI]: 0.25–0.62) in Q2, 0.45 (95% CI: 0.28–0.70) in Q3, and 0.30 (95% CI: 0.19–0.47) in Q4 (Fig. 1a, Table 2). Taken together, the prevalence of diabetic retinopathy, nephropathy, and peripheral neuropathy decreased significantly with increasing TITR quartiles. Furthermore, participants who achieved TITR >50% had a significantly lower prevalence of microvascular complications than those with TITR ≤50% (all P < 0.001; Fig. 1b).

Table 2.

Odds Ratios for Incident Microvascular Complications according to Time in Tight Range Quartiles

	TITR quartiles
Variables	Q1≤42.6%	Q2>42.6 to ≤61.2%	Q3>61.2 to ≤73.4%	Q4>73.4%
N	250	250	247	252
Any microvascular complications
n (%) with outcome	176 (70.4)	128 (51.2)	118 (47.8)	89 (35.3)
Unadjusted ORs (95% CIs)	1.00	0.44 (0.31, 0.64)	0.38 (0.27, 0.56)	0.23 (0.16, 0.33)
Adjusted ORs (95% CIs)^a	1.00	0.39 (0.25, 0.62)	0.45 (0.28, 0.70)	0.30 (0.19, 0.47)
Adjusted ORs (95% CIs)^b	1.00	0.44 (0.27, 0.71)	0.51 (0.30, 0.86)	0.37 (0.21, 0.65)
Diabetic retinopathy
n (%) with outcome	76 (30.4)	63 (25.2)	46 (18.6)	37 (14.7)
Unadjusted ORs (95% CIs)	1.00	0.77 (0.52, 1.14)	0.52 (0.34, 0.80)	0.39 (0.25, 0.61)
Adjusted ORs (95% CIs)^a	1.00	0.75 (0.48, 1.18)	0.66 (0.41, 1.06)	0.50 (0.30, 0.83)
Adjusted ORs (95% CIs)^b	1.00	1.00 (0.62, 1.63)	1.07 (0.62, 1.85)	0.89 (0.48, 1.66)
Diabetic nephropathy
n (%) with outcome	100 (40.0)	66 (26.4)	63 (25.5)	41 (16.3)
Unadjusted ORs (95% CIs)	1.00	0.54 (0.37, 0.79)	0.51 (0.35, 0.75)	0.29 (0.19, 0.44)
Adjusted ORs (95% CIs)^a	1.00	0.73 (0.40, 1.34)	0.78 (0.42, 1.47)	0.48 (0.24, 0.93)
Adjusted ORs (95% CIs)^b	1.00	0.77 (0.41, 1.48)	0.84 (0.42, 1.69)	0.55 (0.25, 1.22)
Diabetic peripheral neuropathy
n (%) with outcome	105 (42.0)	70 (28.0)	58 (23.5)	53 (21.0)
Unadjusted ORs (95% CIs)	1.00	0.54 (0.37, 0.78)	0.42 (0.29, 0.62)	0.37 (0.25, 0.55)
Adjusted ORs (95% CIs)^a	1.00	0.54 (0.35, 0.81)	0.53 (0.34, 0.81)	0.51 (0.33, 0.80)
Adjusted ORs (95% CIs)^b	1.00	0.61 (0.39, 0.94)	0.63 (0.39, 1.03)	0.67 (0.39, 1.15)

Adjusted for age, sex, BMI, duration of diabetes, systolic blood pressure, total cholesterol, HDL cholesterol, log-transformed triglycerides, eGFR, uric acid, log-transformed urine albumin-to-creatinine ratio, smoking, alcohol consumption, use of insulin therapy, use of ACE inhibitors and/or angiotensin II receptor blockers, and use of statin.

Adjusted for above variables plus HbA1c.

CI, confidence interval; ORs, odds ratios.

FIG. 1.

Distribution of microvascular complications across TITR quartiles and TITR ≤50% versus >50%. Data represent the percentage of participants with microvascular complications across TITR quartiles (Q1–Q4) (a) and TITR ≤ 50% versus >50% (Q1–Q2) (b). TITR, time in tight range (70–140 mg/dL).

Each 10% increase in TITR was associated with a 21.9% reduction in the odds of microvascular complications (OR = 0.781; 95% CI: 0.734–0.830; P < 0.001). Similar reductions were observed in the odds of diabetic retinopathy (15.0% reduction; OR = 0.850; 95% CI: 0.797–0.907; P < 0.001), diabetic nephropathy (18.1% reduction; OR = 0.819; 95% CI: 0.770–0.872; P < 0.001), and diabetic peripheral neuropathy (16.0% reduction; OR = 0.840; 95% CI: 0.790–0.892; P < 0.001), as shown in the unadjusted models (Fig. 2). Even after adjustment for potential confounders, TITR remained independently associated with all microvascular complications: any complication (OR = 0.815; 95% CI: 0.755–0.879; P < 0.001), retinopathy (OR = 0.867; 95% CI: 0.800–0.939; P < 0.001), nephropathy (OR = 0.897; 95% CI: 0.809–0.995; P < 0.001), and peripheral neuropathy (OR = 0.875; 95% CI: 0.814–0.940; P < 0.001) (Fig. 2, adjusted model).

FIG. 2.

Impact of TITR on the risk of microvascular complications. Data are odds ratios (ORs) and P values per 10% increase in TITR. The models were adjusted for age, sex, BMI, duration of diabetes, systolic blood pressure, total cholesterol, high-density lipoprotein cholesterol, log-transformed triglycerides, eGFR, uric acid level, log-transformed urine albumin-to-creatinine ratio, smoking, alcohol consumption, use of insulin therapy, use of ACE inhibitors and/or angiotensin II receptor blockers, and use of statins (plus HbA1c).

The addition of HbA1c to the multivariable model resulted in weakening or loss of the TITR–microvasculopathy association, but the overall trend remained unchanged (Table 2 and Fig. 2). This pattern is consistent with the strong inverse correlation between TITR and HbA1c (r = −0.70).

ROC curve analysis identified 54.3% as the optimal TITR cutoff value for the identification of microvascular complications (AUC = 0.647; 95% CI: 0.613–0.681; Fig. 3). The AUC of TITR was comparable to that of HbA1c (AUC = 0.630; 95% CI: 0.596–0.664) in the univariate analysis.

FIG. 3.

Receiver operating characteristic (ROC) curves of TITR for discriminating any diabetes-related microvascular complications (a), retinopathy (b), nephropathy (c), and peripheral neuropathy (d).

Interaction analyses showed no significant effect for insulin therapy on the associations (all P > 0.3). However, diabetes duration significantly modified the associations between TITR quartiles and microvascular complications (P = 0.002 for any microvascular complication). The inverse association between TITR and complications was strongest in participants with shorter duration (<10 years) and weakest in those with longer duration (≥20 years) (Table 3).

Table 3.

Odd Ratios for Incident Microvascular Complications according to Time in Tight Range Quartiles Among Subpopulations

	TITR quartiles
Variables	Q1≤42.6%	Q2>42.6 to ≤61.2%	Q3>61.2 to ≤73.4%	Q4>73.4%	P for interaction
Any microvascular complications
Duration of diabetes, years					0.002
<10	1.00	0.15 (0.06, 0.35)	0.12 (0.05, 0.28)	0.08 (0.03, 0.18)
≧10 to <20	1.00	0.46 (0.23, 0.90)	0.58 (0.28, 1.22)	0.43 (0.20, 0.90)
≧20	1.00	2.13 (0.73, 6.20)	4.69 (1.35, 16.36)	2.01 (0.60, 6.71)
Use of insulin					0.369
Yes	1.00	0.22 (0.06, 0.83)	0.19 (0.04, 0.96)	0.03 (0.00, 0.30)
No	1.00	0.40 (0.24, 0.67)	0.48 (0.29, 0.79)	0.33 (0.20, 0.55)
Diabetic retinopathy
Duration of diabetes, years					0.628
<10	1.00	0.83 (0.30, 2.33)	0.37 (0.13, 1.05)	0.33 (0.12, 0.91)
≧10 to <20	1.00	0.62 (0.31, 1.22)	0.77 (0.37, 1.61)	0.50 (0.22, 1.14)
≧20	1.00	1.15 (0.48, 2.74)	1.01 (0.38, 2.73)	0.85 (0.28, 2.59)
Use of insulin					0.747
Yes	1.00	0.74 (0.30, 1.86)	0.76 (0.24, 2.46)	0.22 (0.03, 1.51)
No	1.00	0.67 (0.39, 1.15)	0.60 (0.34, 1.04)	0.49 (0.28, 0.85)
Diabetic nephropathy
Duration of diabetes, years					0.022
<10	1.00	0.39 (0.13, 1.13)	0.29 (0.11, 0.83)	0.12 (0.04, 0.37)
≧10 to <20	1.00	1.10 (0.40, 3.04)	1.49 (0.48, 4.63)	1.23 (0.35, 4.29)
≧20	1.00	1.05 (0.22, 5.14)	1.49 (0.26, 8.59)	2.61 (0.43, 15.96)
Use of insulin					0.722
Yes	1.00	0.97 (0.30, 3.16)	0.92 (0.20, 4.26)	1.91 (0.26, 14.08)
No	1.00	0.62 (0.31, 1.24)	0.64 (0.32, 1.28)	0.41 (0.20, 0.83)
Diabetic peripheral neuropathy
Duration of diabetes, years					0.046
<10	1.00	0.21 (0.08, 0.55)	0.28 (0.12, 0.65)	0.23 (0.10, 0.54)
≧10 to <20	1.00	0.49 (0.26, 0.91)	0.55 (0.28, 1.09)	0.73 (0.36, 1.47)
≧20	1.00	1.18 (0.51, 2.78)	1.28 (0.48, 3.39)	1.09 (0.38, 3.14)
Use of insulin					0.319
Yes	1.00	0.58 (0.22, 1.51)	0.46 (0.13, 1.60)	0.11 (0.01, 0.85)
No	1.00	0.52 (0.32, 0.86)	0.59 (0.37, 0.97)	0.61 (0.37, 1.00)

In addition, analyses using TIR quartiles showed similar inverse associations with the prevalence of microvascular complications (Supplementary Table S3, Supplementary Fig. S1a). Furthermore, participants who achieved TIR > 70% had a significantly lower prevalence of microvascular complications than those with TIR ≤ 70% (Supplementary Fig. S1b). The adjusted ORs per 10% increase in TIR for any microvascular complication, retinopathy, nephropathy, and peripheral neuropathy are presented in Supplementary Fig S3.

Discussion

Our study evaluated the clinical utility of CGM-derived TITR in patients with T2DM by analyzing its association with the prevalence of microvascular complications, including retinopathy, nephropathy, and peripheral neuropathy, in patients with T2DM and a negative history of cardiovascular disease. Our findings indicated that high TITR quartiles were significantly associated with lower prevalence rates of these complications. Specifically, patients in the highest TITR quartile (TITR > 73.4%) had a 70% lower risk of microvasculopathies, compared to those in the lowest quartile (TITR ≤ 42.6%). Moreover, each 10% increase in the TITR was associated with an 18.5% reduction in the risk of complications. ROC curve analysis determined 54.3% as the optimal TITR cut-off value for the identification of microvascular complications. Our findings extend previous evidence based on TIR by demonstrating comparable associations of TITR with microvascular complications. Therefore, we believe that TITR can potentially serve as a complementary and clinically practical metric for the assessment of near-normoglycemic control. Our study also provides a clinically applicable TITR threshold value that can facilitate the use of this parameter in clinical practice.

As mentioned above, TITR is being increasingly recognized as a valuable endpoint in clinical trials. However, there remains limited evidence linking TITR to diabetes-related outcomes, especially in patients with T2DM. Most previous studies on TITR focused on patients with T1DM. Among them, De Meulemeester et al.¹⁴ reported a negative relationship between TITR and microvascular complications and cerebrovascular disease. In contrast, Wang et al.¹⁵ found that a lower TITR was independently associated with an increased risk of incident diabetic retinopathy. While this association remained significant among participants with well-controlled TIR (>70%), this does not necessarily indicate the superiority of TITR, since TITR and TIR are highly correlated. Rather, these findings suggest that TITR can potentially serve as a complementary parameter, offering additional granularity in the assessment of near-normoglycemic control. Consistent with these findings, two recent analyses of the Virtual DCCT dataset by Lobo et al.¹⁹ and Horton et al.²⁰ demonstrated that both TITR and TIR were significantly associated with the progression of microvascular and macrovascular complications in type 1 diabetics. Our findings further strengthen the notion that CGM-derived metrics reflecting time spent in near-normoglycemic ranges are closely related to long-term diabetes-related vascular outcomes.

In the present study, adjustment for HbA1c weakened the inverse association between TITR and microvascular complications, indicating attenuation of such an association when mean glycemia is accounted for. This attenuation was expected, as TITR and HbA1c share substantial information on average glycemia (r = –0.70). Similar findings were reported by De Meulemees et al.,¹⁴ who demonstrated attenuation of the associations of both TITR and TIR with microvascular complications after adjustment for HbA1c. Considered together, these results suggest that TITR and HbA1c capture overlapping but distinct aspects of glycemic control. Nevertheless, TITR seems to provide clinically useful and complementary insight into glucose stability and near-normoglycemia. We believe that the combined evaluation of HbA1c and TITR can potentially offer a more comprehensive assessment of glycemic quality and microvascular risk.

Our findings are consistent with previous reports and demonstrate that TITR is independently associated with not only diabetic retinopathy but also with diabetic nephropathy and peripheral neuropathy in T2DM. Furthermore, we demonstrated that individuals who achieved the previously proposed TITR target of >50%²¹ had a significantly lower prevalence of microvascular complications, which further supports the validity of this clinical target.²² Moreover, ROC analysis identified 54.3% as the optimal TITR cutoff value for the identification of these complications. These results highlight the importance of maintaining blood glucose levels within a near-normoglycemic range as well as the clinical utility of TITR in minimizing the risk of microvascular complications.

The significant interaction between TITR and diabetes duration indicates that the beneficial association with tight glycemic control is particularly evident in the early stages of diabetes, when microvascular injury is potentially reversible. In contrast, among diabetics with long-standing disease, structural vascular damage may limit the impact of glycemic stability. These findings underscore the importance of achieving early and stable glycemic control to prevent microvascular complications.

Previous studies reported a strong correlation between TITR and TIR¹³; moreover, TITR values have been reported to be consistently 20%–25% lower than the corresponding TIR values.²² Therefore, our findings indicating comparable discriminative performance of TITR and TIR for microvascular complications are consistent with previous reports on T1DM.^13,14,23 Admittedly, the relationship between TITR and TIR is not perfectly linear; specifically, the TITR/TIR ratio tends to increase with improved glycemic control.²² Dunn et al.⁸ suggested that TIR becomes less sensitive to changes as glycemia approaches the normal range, while TITR on the other hand may provide superior risk discrimination. Furthermore, Kim et al.²⁴ recently reported that TITR outperformed TIR in predicting the achievement of HbA1c targets of <6.5% or <7.0%. Taken together, these findings suggest that TITR and TIR should be used in a complementary manner, with TITR serving as a better clinical utility in patients with well-controlled disease.

The strengths of this study include the use of baseline data from a large multicenter prospective cohort, which allowed robust adjustment for multiple confounding factors to evaluate the independent associations of TITR and TIR with diabetes-related microvasculopathies. However, our study has certain limitations. First, since this was a cross-sectional analysis, we could not establish a causal relationship between TITR and the development of complications, and reverse causality could not be entirely ruled out. Accordingly, longitudinal analysis of an ongoing follow-up study¹⁶ is warranted. Second, the study population exclusively comprised Japanese patients with T2DM, which may limit the generalizability of our findings. Third, CGM data were collected over a maximum of 8 days, which may not fully represent long-term glycemic patterns. Although 2–3 days of CGM data have been recently shown to reliably capture intraday glycemic variability,²⁵ longer monitoring periods may improve the robustness of TITR as a representative metric. Fourth, all participants had T2DM and were generally at a low risk for hypoglycemia; therefore, these findings may not be applicable to populations with a higher hypoglycemia risk. Finally, since the mean BMI of our subjects (24.6 kg/m²) was lower than that typically observed in Western populations with T2DM, our findings should be interpreted with caution when extrapolated to populations with higher BMI values.

Conclusions

In conclusion, the present study demonstrated that both TITR and TIR, as assessed using CGM, were inversely associated with the prevalence of microvascular complications in patients with type 2 diabetes. Consistent with previous studies, our findings confirm that TITR, which correlates significantly with TIR, provides comparable information on the quality of glycemic control and can be potentially used as a complementary metric in daily clinical practice. Furthermore, our analysis identified 54.3% as the optimal TITR cut-off value associated with the presence of microvascular complications. Our findings emphasize the clinical utility of TITR as a glycemic metric in future management strategies aimed at achieving near-normoglycemia.

Authors’ Contributions

All authors contributed to the study design and collection of clinical data. Keiichi T. drafted the article. M.G. coordinated and supervised data analysis. Keiichi T., Y.O., T.M., Kenichi T., S.N., N.K., H.Y., K.N., R.I., M.G., I.S., and H.W. collected, analyzed, and interpreted the data; reviewed and edited the article; and approved the final version of the article. Y.O. and H.W. are the principal guarantors of this work; they have full access to all study data and take responsibility for the integrity of the data and the accuracy of data analysis. All authors have read and agreed to the publication of this article.

Footnotes

Acknowledgments

The authors thank the investigators listed in the and all participants for their contributions to this study. The authors also thank D. Takayama, H. Yamada (Soiken Holdings Inc., Tokyo, Japan), and N. Sakaguchi (University of Occupational and Environmental Health, Kitakyushu, Japan).

Author Disclosure Statement

All authors declare no conflicts of interest.

Funding Information

This study was supported by the Japan Agency for Medical Research and Development (AMED) (to H.W.) and the Manpei Suzuki Diabetes Foundation (to H.W.).

Supplemental Material

References

Diabetes

, Nathan

DM.

, et al.; Complications Trial Research G. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med, 1993; 329(14):977–986.

Welsh

, Kirkman

, Sacks

. Role of glycated proteins in the diagnosis and management of diabetes: Research gaps and future directions. Diabetes Care, 2016; 39(8):1299–1306.

Battelino

, Danne

, Bergenstal

, 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.

, Ma

, Zhou

, 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.

Mayeda

, Katz

, Ahmad

, et al. Glucose time in range and peripheral neuropathy in type 2 diabetes mellitus and chronic kidney disease. BMJ Open Diabetes Res Care, 2020; 8(1):e000991.

, Wang

, Shen

, 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.

Wakasugi

, Mita

, Katakami

, et al. Associations between continuous glucose monitoring-derived metrics and diabetic retinopathy and albuminuria in patients with type 2 diabetes. BMJ Open Diabetes Res Care, 2021; 9(1):e001923.

Dunn

, Ajjan

, Bergenstal

, Xu

. 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.

Castaneda

, Arrieta

, van den Heuvel

, et al. 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.

10.

Battelino

, Bergenstal

, Rodriguez

, et al. Efficacy of once-weekly tirzepatide versus once-daily insulin degludec on glycaemic control measured by continuous glucose monitoring in adults with type 2 diabetes (SURPASS-3 CGM): A substudy of the randomised, open-label, parallel-group, phase 3 SURPASS-3 trial. Lancet Diabetes Endocrinol, 2022; 10(6):407–417.

11.

Battelino

, Alexander

, Amiel

, et al. Continuous glucose monitoring and metrics for clinical trials: An international consensus statement. Lancet Diabetes Endocrinol, 2023; 11(1):42–57.

12.

Shah

, DuBose

, Li

, et al. Continuous glucose monitoring profiles in healthy nondiabetic participants: A multicenter prospective study. J Clin Endocrinol Metab, 2019; 104(10):4356–4364.

13.

Shah

, Kanapka

, Akturk

, 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.

14.

De Meulemeester

, Charleer

, Visser

, et al. 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(8):1527–1535.

15.

Wang

, Lu

, Yu

, 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.

16.

Mita

, Katakami

, Okada

, et al. Protocol of a prospective observational study on the relationship between glucose fluctuation and cardiovascular events in patients with type 2 diabetes. Diabetes Ther, 2019; 10(5):1565–1575.

17.

Matsuo

, Imai

, Horio

, et al.; Collaborators developing the Japanese equation for estimated GFR. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis, 2009; 53(6):982–992.

18.

Boscari

, Galasso

, Acciaroli

, et al. Head-to-head comparison of the accuracy of Abbott FreeStyle Libre and Dexcom G5 mobile. Nutr Metab Cardiovasc Dis, 2018; 28(4):425–427.

19.

Lobo

, Kanapka

, Kovatchev

, et al. The Association of Time-in-Range and Time-in-Tight-Range with Retinopathy Progression in the Virtual Diabetes Control and Complications Trial Continuous Glucose Monitoring Dataset. Diabetes Technol Ther, 2025 Jul;27(7):558–561.

20.

Horton

, Kovatchev

, Kanapka

, Beck

. The Virtual DCCT #3: Relationship of HbA1c and CGM Metrics with Cardiovascular Outcomes. Diabetes Technol Ther. 2025 Aug 13.

21.

Akturk

, Battelino

, Castaneda

, et al. 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.

22.

Beck

, Raghinaru

, Calhoun

, Bergenstal

. 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.

23.

, Home

, Zhou

. 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.

24.

Kim

, Park

, Lee

, et al. Time in tight range and time in range for predicting the achievement of typical glucose management indicator and HbA(1c) targets. Diabetologia, 2025; 68(8):1674–1679.

25.

Foreman

, Brouwers

, 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.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.53 MB

0.00 MB