Increased 24-h blood pressure and arterial stiffness in young adults with childhood-onset type 1 diabetes: the Norwegian Atherosclerosis and Childhood Diabetes study

Abstract

Objective:

Individuals with type 1 diabetes (T1D) have increased mortality from cardiovascular disease (CVD) compared to general population, where hypertension is a major risk factor. The aim of this study was to evaluate blood pressure (BP) and arterial stiffness in young adults with childhood-onset T1D compared to healthy control subjects.

Methods:

The atherosclerosis and childhood diabetes (ACD) study is a prospective population-based cohort study, with follow-up every fifth year on early atherosclerosis development and other cardiovascular risk factors in childhood-onset T1D compared to healthy controls. At baseline the cohort was recruited among children with T1D in a defined health-region in Norway along with their friends as healthy control subjects. The original T1D cohort was found to be representative of all children with T1D in Norway at that time. At the 10-year follow-up a cross-sectional analysis of 24-h ambulatory blood pressure measurements (ABPM), including arterial stiffness, was performed, along with an assessment of other CVD risk factors.

Results:

One hundred and twenty-three participants were included in the analysis (T1D n = 82, controls n = 41). The T1D group had mean ± SD age of 24 ± 3 years, diabetes duration of 15 ± 4 years and present HbA1c of 64 ± 14 mmol/mol (8.0 ± 1.3%). None of the T1D participants had previously known hypertension or were on antihypertensive treatment. Females with T1D had significantly higher mean weight, BMI, waist circumference, LDL-cholesterol, and apolipoprotein B than female controls. In females, almost all parameters of ABPM were significantly increased compared to their controls, whereas in males only nighttime blood pressure (BP) was significantly increased compared to controls. In the (stage 2) hypertensive group with T1D, LDL-c (3.1 ± 0.08; ρ = 0.038) was significantly increased. Arterial stiffness was significantly increased in the T1D group (24-h PWV was 8.0 ± 1.0 m/s vs. 7.2 ± 0.6 m/s; ρ = 0.011) compared to controls.

Conclusion:

Young adults with childhood-onset T1D had significantly increased BP and arterial stiffness compared to controls, especially females.

INTRODUCTION

Type 1 diabetes mellitus (T1D) is a chronic and lifelong disease with increased mortality and morbidity compared to the general population [1]. A Swedish register-based study showed that individuals with T1D have a 10–15 year loss of lifespan, mostly due to cardiovascular disease (CVD), where individuals diagnosed in early childhood and females are mostly affected [2]. Recently, the Global Cardiovascular Risk Consortium harmonized individual-level data from >100 studies in 34 countries, where they identified five modifiable cardiovascular risk factors. They demonstrated that systolic blood pressure (SBP) is the most important modifiable risk factor for CVD in the general population [3]. The prevalence of hypertension has been shown to be as high as 6–27% in young adults with T1D, commencing earlier and subsequently becoming more treatment resistant than in the general population [4,5]. Individuals with T1D have a significantly increased risk of coronary artery disease and cerebral stroke, and a strong association with hypertension has been extensively demonstrated [6–8]. Therefore, early targeted therapy is crucial for the prevention of overall CVD in T1D patients.

Arterial stiffness is also an important predictor of future CVD events and improves the risk stratification models [9]. A good indicator of future CVD, especially borderline hypertension, is arterial stiffness assessed using pulse wave velocity (PWV) [10,11]. Studies have shown that arterial stiffness is greater in young patients with T1D than in controls, although with some heterogeneity [12,13].

Despite the knowledge that hypertension and arterial stiffness develop earlier and more extensively in individuals with T1D, there is limited knowledge on sex differences and their association with risk factors. In the present study of the Atherosclerosis and childhood diabetes study, we aimed to evaluate the degree to which blood pressure and arterial stiffness differ between young Norwegian adults with T1D and healthy controls, as well as the associations with other CVD risk factors, with an emphasis on differences between the sexes. To the best of our knowledge, no similar study has been conducted in Norway.

METHODS

Study design

This was a cross-sectional analysis based on data from the 10-year follow-up of the prospective Norwegian Atherosclerosis and Childhood Diabetes (ACD) study, but it included total exposure of certain risk factors e.g. HbA1c over a predefined time in childhood and adolescences. The cohort consisted of individuals with T1D and healthy controls with similar age and sex distributions. Clinical examination and blood sampling were conducted at the Oslo University Hospital, Ullevål, Oslo, Norway.

Study population

At baseline, 314 individuals with T1D and 120 controls aged 8–18 years were included in the ACD study; the two groups had similar age and socioeconomic background. The cohort was recruited among children and adolescents with T1D from the defined Southeast Health Region of Norway, covering one large city (Oslo), smaller cities, towns and countrysides, that were registered in the Norwegian Childhood Diabetes Registry (NCDR). The control group was recruited among friends of the participating T1D individuals. The T1D patients recruited to the study at baseline were compared with T1D patients in the rest of the country registered in the NCDR and were found to be representative of the childhood T1D population in the whole country. Details of the recruitment process and examinations have been previously described [14].

At the five-year follow-up, 53 new controls and 15 new individuals with T1D were included, thus expanding the entire cohort to 502 participants (329 with T1D and 173 controls). At the ten-year follow-up, which was conducted between 2017–2019, 59% of the total cohort returned. The reasons for dropping out of the study have been explained and discussed in our previous papers [15].

Clinical examination and risk factor variables

Fasting venous blood samples were collected in the morning and urine samples were collected at every follow-up. The clinical examinations included height, weight, and waist circumference measurements in a standardized manner. The following information was collected via interviews with participants: medical history, current medications (including insulin regimens), and microvascular complications of diabetes.

Norwegian Childhood Diabetes Registry

The NCDR is a national quality registry that collects annual clinical data from children with T1D from diagnosis until the age of 18 years, upon providing consent [16]. At baseline, we compared our T1D participants with all children and adolescents in the registry and found our T1D cohort to be representative of the entire childhood diabetes population in Norway at that time. During the ten-year follow-up, we extracted annual supplementary data for individuals with T1D from the registry, covering the years 2005–2012. We used the NCDR data to perform data analysis with the area under the curve (AUC) for known risk factors and covariates, such as total cholesterol, HDL-cholesterol (HDL-c), LDL-cholesterol (LDL-c), apolipoprotein B (ApoB), body mass index (BMI), and HbA1c.

Laboratory analysis

The serum glucose, creatinine, and lipid profiles were measured using standard methods. HbA1c was determined by high-performance liquid chromatography (Variant; Bio-Rad, Richmond, CA, USA) at a central DCCT-standardized laboratory (the normal reference range was 20–41 mmol/mol, and the intra-assay coefficient of variation was <3%) at the Clinical Chemistry Department of Oslo University Hospital, Oslo, Norway.

Office blood pressure and ambulatory blood pressure measurement

Office blood pressure was measured using a cuff on the nondominant upper arm after approximately 20 min of rest and overnight fasting. The participants were measured three times, and the average of these measurements was recorded.

ABPM monitoring was performed using the Tensiomed Arteriograph 24 (Budapest, Hungary). Arteriograph 24 measures peripheral systolic (S) and diastolic (S) BP over 24-h, central (aortic) SBP, and heart rate. It also measures arterial stiffness by measuring the central pulse wave velocity (PWV), as well as the brachial and central augmentation index (AIx). The method and device used in this study being evaluated and described in other publications [16,17,18,19].

Ambulatory BP was recorded every 20 min during the day and every 30 min during the night, based on individual sleep cycles. The criteria for inclusion in this analysis were that ≥70% of the expected measurements were successfully recorded separately for blood pressure measurements and arterial stiffness parameters. Hypertension was defined based on the 2024 European Society of Cardiology Guidelines (ESC), the American Diabetes Association (ADA) clinical practice recommendations and ACC/AHA blood pressure thresholds [19,20]. Office hypertension was defined as blood pressure ≥130/80 [20,21]. For 24-h ABPM, ACC/AHA thresholds were used, with stage 2 hypertension defined as ≥130/80 mmHg for the 24-h average mean, ≥135/85 mmHg for daytime BP and ≥120/70 mmHg for nighttime BP [20]. Nocturnal dipping of systolic blood pressure was also assessed, where dipping < 10% was defined as absent, dipping 10–20% as normal, and dipping >20% as excessive [19].

Statistical analysis

Demographic and clinical data are presented as mean ± standard deviation (SDs) for normally distributed data and medians with minimum and maximum values for nonnormally distributed data. Differences in continuous variables between the T1D and control groups were tested using the independent sample t-test for normally distributed data or the Mann–Whitney U test for skewed data. Correlation analyses between continuous variables were performed using Pearson's correlation coefficient (R). The chi-square test for contingency tables or Fisher's exact test was used to detect differences in the categorical variables. Analyses were performed on the total study population and were stratified according to sex.

Glycemic exposure over time was calculated via trapezoidal integration of the area under the curve (AUC) using annual data from the NCDR (2005–2017) and including data from all three examinations of the ACD study. We calculated the total area under the curve-HbA1c curve (tAUC_HbA1c). Finally, we repeated the analysis described above for t_AUC for total cholesterol, LDL-c, HDL-c, and BMI.

Multiple linear regression analyses were performed to investigate the risk factors associated with either daytime or nighttime SBP, DBP, and central SBP as outcome variables within the type 1 diabetes group. Possible risk factors for the outcomes were selected from the literature. The following factors were included: age, diabetes duration, sex, smoking status, ln urine albumin-creatinine ratio (ln u-ACR), ln triglyceride (lnTg), HbA1c, LDL-c, HDL-c, and BMI. In addition, heart rate and central SBP were included in the analysis of PWV. We also analyzed how the different risk factors differ between the male and female T1D participants. Only the significant results are presented as regression coefficients (B) with 95% confidence intervals (CIs) and explained variances (R²).

A significance level of 5% was considered statistically significant. Statistical analyses were performed using IBM SPSS statistics (version 29.0; IBM SPSS Inc., Armonk, NY, IBM Corp.) and STATA 18.0 (StataCorp LP, College Station, TX, USA).

Ethical declarations

Written informed consent was obtained from all participants and parents of those younger than 18 years at baseline and at every follow-up. The Norwegian Regional Committee for Research Ethics (REK) approved the study protocol (REK id 2011/1818). This study was conducted in accordance with the principles of the Declaration of Helsinki.

Data statement

All data generated or analyzed during this study are included in this published article and its supplementary information files.

RESULTS

At the 10-year follow-up, 185 (92%) participants agreed to undergo ABPM. Among 185 individuals who underwent ABPM, 123 (66%) completed sufficient measurements and were included in this analysis: 82 individuals with T1D (67%) and 41 controls (33%).

Table 1 presents an overview of participants’ characteristics. Male participants with T1D (35%) were 24.3 ± 2.7 years old and had a duration of diabetes of 15.2 ± 4.4 years. Female participants with T1D (65%) were 24.6 ± 2.9 years, with a duration of diabetes of 15.5 ± 3.4 years. The mean HbA1c levels in the group with T1D were 61 ± 9 mmol/mol (7.7 ± 0.8%) and 65 ± 17 mmol/mol (8.1 ± 1.5%) in males and females, respectively. All participants with T1D received intensive insulin treatment and the majority used insulin pumps (68%). Compared to their controls, females with T1D presented significantly greater weight, BMI, waist circumference, number of obese, LDL-c, ApoB, and HbA1c levels. Only HbA1c was significantly increased in male participants with T1D compared to controls. Between the T1D sexes height, weight, waist circumference, total cholesterol, HDL-c, ApoB, ApoA1, eGFR, and u-ACR were significantly different (Table 1 (S1), Supplementary Digital Content).

TABLE 1.

Demographic and descriptive data of the participants at the 10-year follow-up

	Total group			Male group			Female group
Variable	T1D	Controls	ρ	T1D	Controls	ρ	T1D	Controls	ρ
N	82	41		29	10		53	31
Age (years)	24.2 ± 2.8	23.8 ± 3.2	ns	24.3 ± 2.7	24.1 ± 3.7	ns	24.6 ± 2.9	23.8 ± 3.1	ns
Age at diagnosis date (years)	8.2 ± 3.7	–		8.3 ± 3.9	–		8.1 ± 3.6	–
Duration of diabetes (years)	15.4 ± 3.7	–		15.2 ± 4.4	–		15.5 ± 3.4	–
Insulin pump usage (%)	56 (68%)	–		22 (76%)	–		34 (64%)	–
Smokers	3	6		2	2		1	4
Blood pressure medications	0	1		0	0		0	1
Statin therapy	3	0		1	0		2	0
Height (cm)	173.5 ± 9.6	171.7 ± 8.3	ns	184.2 ± 5.1	181.6 ± 5.3	ns	167.7 ± 5.7	168.5 ± 6.4	ns
Weight (kg)	74.9 ± 14.2	67.3 ± 13.4	0.005	82.3 ± 12.9	80.3 ± 13.4	ns	70.9 ± 13.3	63.1 ± 10.5	<0.001
BMI (kg/m2)	24.8 ± 4.0	22.6 ± 3.3	0.004	24.2 ± 3.8	24.3 ± 3.6	ns	26.3 ± 4.9	22.7 ± 2.9	<0.001
Overweight, BMI (≥25 kg/m2) (%)b	22 (27%)	10 (24%)	ns	7 (24%)	5 (50%)	ns	15 (28%)	5 (16%)	ns
Obese, BMI (≥30 kg/m2) (%)b	12 (15%)	0	0.010	3 (10%)	0	ns	9 (17%)	0	0.023
Waist circumference (cm)	83.9 ± 10.9	77.7 ± 9.7	0.002	87.5 ± 11.2	86.9 ± 10.1	ns	82.0 ± 10.3	74.7 ± 7.6	<0.001
HbA1c (mmol/mol)	64 ± 14	32 ± 3	<0.001	61 ± 9	32 ± 2	<0.001	65 ± 17	32 ± 3	<0.001
HbA1c (%)	8.0 ± 1.3	5.1 ± 0.2	<0.001	7.7 ± 0.8	5.1 ± 0.2	<0.001	8.1 ± 1.5	5.1 ± 0.2	<0.001
Total cholesterol (mmol/l)	4.4 ± 0.9	4.3 ± 0.7	ns	4.0 ± 0.7	4.2 ± 0.6	ns	4.6 ± 0.9	4.3 ± 0.7	ns
HDL-c (mmol/l)	1.6 ± 0.4	1.6 ± 0.5	ns	1.4 ± 0.3	1.4 ± 0.3	ns	1.7 ± 0.4	1.7 ± 0.5	ns
LDL-c (mmol/l)	2.6 ± 0.7	2.5 ± 0.6	ns	2.4 ± 0.7	2.7 ± 0.6	ns	2.7 ± 0.7	2.4 ± 0.6	0.006
Triglycerides (mmol/l)a	0.8 (0.3, 6.1)	0.8 (0.3, 2.7)	ns	0.8 (0.3, 3.5)	0.9 (0.4, 2.7)	ns	0.8 (0.3, 6.1)	0.8 (0.3, 2.5)	ns
Apolipoprotein B (g/l)	0.8 ± 0.2	0.8 ± 0.1	ns	0.8 ± 0.2	0.8 ± 0.2	ns	0.9 ± 0.2	0.7 ± 0.1	0.013
Apolipoprotein A1 (g/l)	1.6 ± 0.3	1.6 ± 0.3	ns	1.4 ± 0.3	1.4 ± 0.3	ns	1.7 ± 0.3	1.6 ± 0.3	ns
Apo B-to-ApoA1 ratio	0.5 ± 0.1	0.5 ± 0.1	ns	0.5 ± 0.1	0.5 ± 0.1	ns	0.5 ± 0.1	0.4 ± 0.1	ns
Creatinine (μmol/l)	68.5 ± 11.9	69.1 ± 11.8	ns	76.5 ± 10.0	80.1 ± 11.8	ns	63.9 ± 10.4	66.1 ± 10.0	ns
eGFR	116.9 ± 13.4	112.7 ± 14.7	ns	121.5 ± 9.6	117.8 ± 13.4	ns	114.4 ± 14.5	111.1 ± 15.0	ns
u-ACR (mg/mmol)a	0.5 (0.1, 232.6)	0.3 (0.1, 93.4)	ns	0.4 (0.1, 7.7)	0.2 (0.1, 0.4)	0.053	0.6 (0.2, 232.6)	0.4 (0.1, 93.4)	ns

Data are presented as mean ± standard deviations (SDs), analyzed by independent samples t-test.

^aPresented as median (minimum and maximum) by Mann–Whitney U test.

^bNumbers and percentages by Pearson's chi-square test or Fisher's exact test as appropriate.

The differences in ABPM between the T1D group and the controls are highlighted in Table 2 (S2), Supplementary Digital Content, where almost all measurements were significantly increased in participants with T1D. After stratification according to sex, females with T1D had significantly greater means of most BPs than their controls. Compared to their controls, males with T1D had significantly increased office DBP and all nocturnal BP measurements (S1). Between the T1D sexes, significant differences were observed in 24-h and daytime pulse pressure, as well as 24-h and nighttime heart rate.

Table 2 illustrates different cutoff values for elevated blood pressure and hypertension based on American and European guidelines and shows significant differences between the T1D group as a whole and controls for most BP thresholds in the ACC/AHA guidelines and the ESC thresholds. However, when the data were stratified according to sex, significant differences were observed only in the female subgroup (Table 2). No significant differences in BP were observed between the female and male T1D groups (data not shown).

TABLE 2.

Hypertension in young adult individuals with type 1 diabetes and controls – total and stratified according to sex

	Total			Male			Female
Variable	T1D	Controls	ρ	T1D	Controls	ρ	T1D	Controls	ρ
Participants	82	41		29	10		53	31
Office BPa
Elevated BP (≥120/80 mmHg)	26 (32%)	4 (10%)	0.007	8 (28%)	1 (10%)	ns	18 (34%)	3 (10%)	0.011
Stage 1 hypertension (≥130/80 mmHg)	10 (12%)	1 (2%)	ns	2 (7%)	1 (10%)	ns	8 (15%)	0	0.023
Stage 2 hypertension (≥140/90 mmHg)	3 (4%)	0	ns	–	–		3 (6%)	0	ns
Blood pressure goal in diabetesb (≥130/80 mmHg)	10 (12%)	1 (2%)	ns	2 (7%)	1 (10%)	ns	8 (15%)	0	0.023
Isolated SBP (≥140 mmHg, DBP ≤90 mmHg)c	7 (9%)	0	ns	4 (14%)	0	ns	3 (6%)	0	ns
ABPMc
24-h Hypertension (≥130/80 mmHg)	11 (13%)	0	0.015	3 (10%)	0	ns	8 (15%)	0	0.024
Daytime hypertension (≥135/85 mmHg)	11 (13%)	1 (2%)	ns	3 (10%)	0	ns	8 (15%)	1 (3%)	ns
Nighttime hypertension (≥120/70 mmHg)	14 (17%)	1 (2%)	0.020	6 (15%)	0	ns	8 (15%)	1 (3%)	ns
ABPMa
24-h Elevated blood pressure (≥115/75 mmHg)	25 (30%)	3 (7%)	0.003	8 (20%)	1 (10%)	ns	17 (32%)	2 (6%)	0.007
24-h stage 1 hypertension (≥125/75 mmHg)	24 (29%)	2 (5%)	0.002	8 (20%)	1 (10%)	ns	16 (30%)	1 (3%)	0.004
24-h stage 2 hypertension (≥130/80 mmHg)	11 (13%)	0	0.015	3 (10%)	0	ns	8 (15%)	0	0.024
Daytime elevated blood pressure (≥120/80 mmHg)	24 (29%)	3 (7%)	0.005	7 (24%)	1 (10%)	ns	17 (32%)	2 (6%)	0.007
Daytime stage 1 hypertension (≥130/80 mmHg)	23 (28%)	3 (7%)	0.009	7 (24%)	1 (10%)	ns	16 (30%)	2 (6%)	0.012
Daytime stage 2 hypertension (≥135/85 mmHg)	11 (13%)	1 (2%)	ns	3 (10%)	0	ns	8 (15%)	1 (3%)	ns
Nighttime elevated blood pressure (≥100/65 mmHg)	30 (37%)	8 (20%)	ns	10 (34%)	1 (10%)	ns	20 (38%)	7 (23%)	ns
Nighttime stage 1 hypertension (≥110/65 mmHg)	30 (37%)	8 (20%)	ns	10 (34%)	1 (10%)	ns	20 (38%)	7 (23%)	ns
Nighttime stage 2 hypertension (≥120/70 mmHg)	14 (17%)	1 (2%)	0.020	6 (21%)	0	ns	8 (15%)	1 (3%)	ns

Data are presented as numbers and percentages, analyzed by Pearson's chi-square or Fisher's exact test as appropriate.

^a2025 ACC/AHA Guidelines.

^b2025 ADA Guidelines.

^c2024 ESC thresholds.

Figure 1 illustrates the BP differences in the total, male, and female T1D groups compared to their controls, and is based on the cutoff threshold from the ADA and 2025 ACC/AHA and 2024 ESC guidelines.

FIGURE 1.

Office blood pressure according to ADA guidelines and ambulatory blood pressure according to ACC/AHA and ESC guidelines, stratified according to sex.

No significant differences in nightly dipping BP were found in the T1D group as a whole or stratified by sex compared to the control group (S4), or between the sexes in T1D (S5). Arterial stiffness was significantly increased in the T1D group in terms of central SBP and PWV at both daytime and nighttime measurements (S6). The group was not stratified by sex because of the small number of male participants.

Associations with different covariates

The BP parameters in the T1D group were adjusted for all registered risk factors, and the significant results of the multiple linear regression analysis are summarized in Table 7 (S7), Supplementary Digital Content. LDL-c was found to be associated with both daytime SBP and daytime and nighttime DBP. Further analysis revealed that u-ACR was significantly associated with nighttime SBP as well as with daytime and nighttime DBP. HbA1c level was significantly associated with nighttime DBP. We also analyzed if the relationship between the risk factors and ABPM differ between male and female T1D groups (S8). We found that HDL-c had significant less effect on nighttime SBP and DBP in females with T1D compared to males.

Significant associations with arterial stiffness as demonstrated by PWV are shown in Table 9 (S9), Supplementary Digital Content, where only u-ACR was significant.

None of the risk factors collected during childhood and adolescence (t_AUC) were significantly associated with office/ambulatory BP, PWV, or stage 2 hypertension measured during the 10-year follow-up (data not shown).

Table 3 presents the clinical differences between T1D individuals with stage 1 and 2 hypertension and normotensive T1D participants. These data indicate that individuals with stage 2 hypertension were younger at diabetes diagnosis, withlonger diabetes duration, and higher mean HbA1c, triglyceride, and median u-ACR values, although these differences were not statistically significant. In contrast, they had significantly increased total cholesterol, LDL-c, and ApoB levels.

TABLE 3.

Clinical parameter differences between individuals with elevated and hypertensive stage 1/hypertensive stage 2 disease and normotensive individuals with T1D

Variable	Elevated and hypertensive stage 1	ρ	Normotensive	ρ	Hypertensive stage 2
N	14		15		11
Age (years)	24.6 ± 3.3	ns	25.1 ± 2.1	ns	24.1 ± 2.6
Age at diagnosis (years)	9.6 ± 4.3	ns	9.1 ± 3.3	ns	6.9 ± 3.5
Diabetes duration (years)	14.7 ± 2.5	ns	15.1 ± 3.8	ns	17.0 ± 3.1
BMI (kg/m2)	24.5 ± 3.9	ns	23.7 ± 2.5	ns	24.6 ± 4.4
Waist circumference (cm)	84.5 ± 13.3	ns	79.5 ± 7.7	ns	83.6 ± 12.1
HbA1c (mmol/mol)	63 ± 12	ns	60 ± 8	ns	75 ± 26
HbA1c (%)	7.9 ± 1.1	ns	7.6 ± 0.8	ns	9.0 ± 2.4
Total Cholesterol (mmol/l)	4.2 ± 0.9	ns	4.1 ± 0.6	0.006	5.1 ± 1.0
HDL-c (mmol/l)	1.8 ± 0.5	ns	1.5 ± 0.3	ns	1.8 ± 0.7
LDL-c (mmol/l)	3.0 ± 1.0	ns	2.5 ± 0.6	0.038	3.1 ± 0.8
Triglycerides (mmol/l)a	0.87 (0.39, 1.23)	ns	0.79 (0.38, 2.37)	ns	1.14 (0.39, 3.50)
Apolipoprotein B (g/l)	0.9 ± 0.2	ns	0.8 ± 0.2	0.038	1.0 ± 0.2
Apolipoprotein A1 (g/l)	1.6 ± 0.3	ns	1.5 ± 0.2	ns	1.8 ± 0.5
Creatinine (μmol/l)	69.3 ± 11.5	ns	69.4 ± 12.1	ns	70.5 ± 9.4
eGFR	117.8 ± 10.4	ns	112.5 ± 17.7	ns	112.0 ± 14.8
u-ACR (mg/mmol)a	0.50 (0.20, 7.70)	ns	0.40 (0.20, 1.80)	ns	0.90 (0.30, 232.60)
PWV (m/s)	8.2 ± 1.3	ns	7.7 ± 0.7	ns	8.1 ± 1.3

The data are presented as the mean ± SDs, analyzed using independent samples t-tests.

^aPresented as the median (minimum and maximum), analyzed by the Mann–Whitney U test.

DISCUSSION

The main findings of the present study were that the mean BP and arterial stiffness parameters were greater in young adults with T1D than in controls, especially in females, and that the hypertensive T1D group presented unfavorable lipid and glycemic profiles. In our study, 30% of T1D subjects had elevated office BP levels compared to 10% in the control group. With ambulatory measurements, 29% of the T1D individuals had stage 1 hypertension, and 13% had stage 2 hypertension, as defined by the ACC/AHA and ESH guidelines [20,22]. In the control group, these proportions were 5% and 0%, respectively. We also demonstrated that the hypertensive T1D group was younger at diagnosis, had a longer diabetes duration, a more unfavorable glycemic burden (HbA1c 75 mmol/mol (9%)), and significantly increased LDL-c and ApoB compared to the normotensive T1D group. Our findings are in concordance with those of previous studies on office BP measurements, which revealed that young individuals with T1D are more likely to have hypertension at a very young age [23,24]. Although only 13% of patients in our study had (stage 2) hypertension, our results are important. Prehypertension (elevated BP and stage 1 hypertension), which is more common in the general young population than (stage 2) hypertension, is a major antecedent for the development of hypertension and CVD later in life [7]. If the disease is diagnosed early, it may normalize more effectively even with lifestyle interventions [7]. According to the findings of the Global Cardiovascular Risk Consortium [3], these patients would most likely benefit from BP-lowering treatment, with their data showing that lower SBP might offer the greatest potential for CVD prevention. Furthermore, in a meta-analysis of 61 prospective studies, 20 mmHg higher SBP and 10 mmHg higher DBP were associated with a doubling of the risk of death from CVD [8]. The 2025 ACC/AHA guidelines ABPM treatment thresholds recommend commencement of treatment at awake ≥130/80 mmHg or average 24-h ≥125/75 mmHg in patients with diabetes, where a higher proportion of our patients (28-29%) would fall within [20].

A report from the SEARCH for Diabetes in Youth study revealed an association between office hypertension and glycemic control [23]. In our study, we found no association between glycemic burden and increased BP in our T1D cohort, despite having estimates of glycemic burden since the time of diagnosis. Analyses from the SWEET international database revealed that the presence of multiple modifiable risk factors is more common in patients with hypertensive T1D [25]. They also reported that age, disease duration, dyslipidemia, and obesity/overweight were major determinants of increased BP or hypertension [25]. In our study, the T1D group with (stage 2) hypertension was younger at diabetes onset and had higher HbA1c levels, although the difference was not significant compared to the normotensive group. However, they did have significantly increased total cholesterol, LDL-cholesterol, and apolipoprotein B levels compared to those in the normotensive T1D group. Multiple regression analyses revealed that LDL-c levels were consistently associated with hypertension parameters. However, no association was found between hypertension and the lipid burden in young individuals. This finding suggests that dyslipidemia does not precede hypertension, but rather coexists in hypertensive individuals with T1D.

We also found significantly increased arterial stiffness in the T1D group compared to controls for most parameters, most significantly for PWV. We previously demonstrated increased PWV assessed by MRI in young individuals with type 1 diabetes compared to controls after 10 years of disease duration [26]. We found an association between albuminuria and PWV, akin to previous findings in large community cohorts [27,28].

With respect to sex, we demonstrated that females with T1D had a more unfavorable metabolic profile and higher blood pressure parameters than their female controls and male counterparts. Other studies have reported similar results in younger women [29,30]. Increased insulin resistance secondary to obesity is presumably a contributing factor, as females tend to have higher BMIs than men do. In the general population, there is a linear relationship between increased weight and hypertension [31]. Although 17% of our females were obese, the average BMI was similar between the stage 2 hypertensive and normotensive groups, and regression analyses revealed no association between hypertension and BMI in our cohort. This finding might indicate that obesity may not be the most important contributor to hypertension in our T1D group, but the group is quite young and the number of those with hypertension small. Additionally, central obesity has been shown to be a better predictor of incidental hypertension in young T1D patients [32], unfortunately we do not have cumulative data on central obesity from diagnosis till now. We did, however, find that HDL-c had significantly less effect on lowering nighttime SBP and DBP in the female T1D group compared to the male T1D group. Other studies have shown that although HDL-c is increased in women with T1D compared to men, the heart-protective effect of HDL-c is reduced in women with T1D [33,34]. We have previously demonstrated that HDL-c function is reduced in T1D [35].

None of our participants with T1D had previously known hypertension or were on antihypertensive treatment. However, our results showed that 13% had (stage 2) hypertension. Thus, masked hypertension may be a real concern in young Norwegian patients with T1D. A recent study by Staplin et al. revealed that nighttime systolic BP was much more informative than office BP for mortality risk in the general population [36]. Nocturnal hypertension is by far the most important predictor of CVD risk in individuals with diabetes [5]. We found that 17% of our individuals with T1D only had nocturnal hypertension, which was also significantly greater than that in the control group, although none had a significant loss of nocturnal dipping. This is an important finding that remains unnoticed when only measuring office BP, supporting the importance of measuring ambulatory BP in T1D individuals on a regular basis to identify patients who might benefit from early intervention to reduce the future risk of CVD. Furthermore, we found a strong association between u-ACR and nighttime BP which is in accordance with findings in young adults with T1D reported in a recent meta-analysis [37]. As ABPM are relatively costly, time-consuming measures that might not be easily applied in every diabetic subject in clinical practice, maybe priority should be given to those who need or might benefit from it the most. Given the strong association between albuminuria and nocturnal hypertension in our study as well as other studies, measuring u-ACR, a less expensive marker, might be used to select those that might benefit the most from ABPM and even to detect patients at highest risk of developing CVD. However, results from some studies suggest that elevated nighttime BP precedes the development of microalbuminuria in T1D [38] and ABPM might therefore be an even earlier and better marker of complications in diabetes.

In conclusion, young adults with childhood-onset T1D in our study had elevated blood pressure and premature arterial stiffness compared to the controls. Thirty percentage had elevated BP and 29% had stage 1 hypertension, as defined by the 2025 ACC/AHA guidelines[20] and would benefit from antihypertensive treatment to reduce CVD risk. Elevated BP and arterial stiffness were associated with other frequent CV risk factors, most notably LDL-c level and albuminuria. These changes were more pronounced in females with T1D who had significantly more unfavorable blood pressure profiles than their controls. Our findings imply that ABPMs should be included in regular clinical assessments of T1D patients from a young age, especially those who are diagnosed early.

Strengths and limitations

The strengths of this study include the young age of the participants, negligible use of antihypertensive or lipid-lowering medication. The main limitation of this study was the small sample size. A recent study has also shown that higher uric acid levels at T1D diagnosis may be linked to micro- and macrovascular complications [39]. Unfortunately, we don’t have current data on uric acid, but we will implement this in future studies. Given the cross-sectional design of the study we are, unfortunately, neither able to reveal causal relationship between potential risk factors and elevated BP or increased stiffness, nor to study any mediators of the relationship between those.

ACKNOWLEDGEMENTS

The authors thank all the adolescents and young adults, as well as their parents, for their participation. The authors also thank Torild Skrivarhaug and Siv Janne Kummernes of the Norwegian Childhood Diabetes Registry. We thank Prof. Ingebjørg Seljeflot for overseeing the biobank of the study and providing us with laboratory technicians. We acknowledge Vibeke Bratseth and Sissel Åkra for the collection of blood samples and laboratory analyses.

Author contributions: A.S. initiated this study. She performed most of the clinical examinations at the 10-year follow-up of the ACD study, analyzed the data, and drafted the manuscript. C.B. contributed to and controlled the statistical analysis. K.D.J. and H.D.M. initiated and designed the ACD study. All authors contributed to the interpretation of the data and participated in the critical revision of the manuscript. All authors approved the final version of the manuscript.

Data availability statement: The original contributions presented in this study are included in the article and have not been published elsewhere. Further inquiries can be directed at the corresponding author.

Source of funding: This work was part of a PhD scholarship funded by the Southeastern Norway Regional Health Authority. Laboratory costs were supported by grants from the Stein Erik Hagen Foundation for Clinical Heart Research, Oslo, Norway. Various running costs are funded by grants from the Oslo Diabetes Research Center, Johan Selmer Kvanes Foundation, and Norwegian Diabetes Foundation.

Conflicts of interest

The authors have no financial or nonfinancial competing interests to declare.