Bromocriptine Improves Central Aortic Stiffness in Adolescents with Type 1 Diabetes: Arterial Health Results from the BCQR-T1D Study

Abstract

Background:

The presence of vascular dysfunction is a well-recognized feature in youth with type 1 diabetes (T1D), accentuating their lifetime risk of cardiovascular events. Therapeutic strategies to mitigate vascular dysfunction are a high clinical priority. In the bromocriptine quick release T1D study (BCQR-T1D), we tested the hypothesis that BCQR would improve vascular health in youth with T1D.

Methods:

BCQR-T1D was a placebo-controlled, random-order, double-blinded, cross-over study investigating the cardiovascular and metabolic impact of BCQR in T1D. Adolescents in the BCQR-T1D study were randomized 1:1 to phase-1: 4-weeks of BCQR or placebo after which blood pressure (BP) and central aortic stiffness measurements by pulse wave velocity (PWV), relative area change (RAC) and distensibility from phase-contrast MRI, were performed. Following a 4-week washout period, phase 2 was performed in identical fashion with the alternate treatment.

Results:

Thirty-four adolescents (mean age 15.9±2.6 years, HbA1c 8.6±1.1%, BMI %ile 71.4±26.1, medianT1D duration 5.8 years) with T1D were enrolled and had MRI data available. Compared to placebo, BCQR therapy decreased systolic [Δ=−5 mmHg, (95%CI: −3, −7), p<0.001) and diastolic BP (Δ=−2 mmHg, (95%CI: −4, 0), p=0.039). BCQR reduced ascending aortic PWV (Δ=−0.4 m/s, p=0.018), and increased RAC (Δ=−2.6%, p=0.083) and distensibility (Δ=0.08 %/mmHg, p=0.017). In the thoraco-abdominal aorta, BCQR decreased PWV (Δ=−0.2 m/s, p=0.007) and increased distensibility (Δ=0.05 %/mmHg, p=0.013).

Conclusions:

BCQR improved BP and central and peripheral aortic stiffness and pressure hemodynamics in adolescents with T1D over 4 weeks vs. placebo. BCQR may improve aortic stiffness in youth with T1D, supporting future longer-term studies.

Graphical Abstract

INTRODUCTION

The incidence of youth-onset type 1 diabetes mellitus (T1D) is increasing annually worldwide and is projected to continuously rise over the next 40 years^1,2. The presence of peripheral vascular dysfunction along with accelerated large arterial stiffness and impaired cardiac function has been clearly demonstrated in youth with T1D, magnifying the risk of cardiovascular events. Indeed, epidemiological studies indicate an association between an earlier age of T1D diagnosis and adverse cardiovascular outcomes^3,4. Therefore, the development of therapeutic strategies to slow progression of systemic vascular disease in youth with T1D is of critical importance.

The recent results from our EMERALD (Effects of MEtformin on caRdiovascular function in AdoLescents with type 1 Diabetes) study indicated that adolescents with T1D already have pathological vascular remodeling as evidenced by increased MRI-assessed aortic stiffness, wall shear stress, and carotid intima-media thickness (cIMT)⁵. Attenuation of the stiffening process in large elastic vessels such as the aorta and its primary branches is of particular interest given the importance of stiffness as an independent marker of all-cause and cardiovascular morbidity and mortality in non-T1D populations, and serves frequently as a therapeutic end-point in clinical trials^6–10. In adults with T1D, central aortic and large artery stiffness has also been shown to predict end-organ disease and peripheral arterial vasculopathy^11,12. Recently, phase-contrast MRI has enabled the non-invasive and simultaneous evaluation of global and regional central aortic stiffness. Furthermore, MRI has been shown to be sensitive to detecting interventional effects in prior trials conducted over relatively short periods of time^5,6.

Bromocriptine is an exogenous sympatholytic D2-dopamine agonist that has been reported to improve insulin sensitivity¹³. Results from animal model studies imply that bromocriptine increases morning dopamine levels in hypothalamic and thalamic nuclei. Increased morning dopamine levels in turn lead to reduced sympathetic nervous system activity, hepatic glucose production, lipolysis, and lipogenesis. Cumulatively this effect improves glucose tolerance and insulin sensitivity^13,14. Recently, the once-daily AM bromocriptine quick release (BCQR) formulation was developed to enhance morning central nervous system dopaminergic activity and has been approved for therapeutic use in adults with type 2 diabetes (T2D). In addition to improving insulin sensitivity, BCQR therapy has been found to decrease sympathetic nervous system and renin angiotensin aldosterone system activity¹³. Importantly, BCQR reduced the incidence of major adverse cardiovascular events in participants with T2D after 1-year of treatment^15,16. However, in T1D, the therapeutic efficacy of BCQR and its effects on cardiovascular health remain unstudied.

Accordingly, we conducted the Bromocriptine in Type 1 Diabetes (BCQR-T1D) study, a 4-week placebo-controlled, random-order, double-blinded, cross-over study to investigate the effects of BCQR as an adjunct therapy to insulin. We previously reported from the BCQR-T1D study that BCQR did not impact most measures of glycemia in youth or adults with T1D, but increased serum creatinine in youth (potentially implying attenuated glomerular hyperfiltration) and lowered systolic and diastolic blood pressure in T1D adults and youth and reduced systemic vascular resistance in adults¹⁷. Here we report a detailed analysis of the large central elastic- as well as medium muscular arteries, all of which are contributors to cardiovascular risk^18–20. Specifically, we investigated BCQR’s effects on central and peripheral vasculature, as measured by endothelial function and central and peripheral arterial stiffness.

METHODS

Collected data, procedural/analytical methods, as well as study materials will be made available to other researchers for purposes of reproducing the results or replicating the procedure. This study was approved by the Colorado Multiple Institutional Review Board, and all participants and guardians provided written informed assent and/or consent as appropriate for age.

Participant and Study Design

As part of the BCQR-T1D study (ClinicalTrials.gov Identifier: NCT02544321), adolescents from the overall BCQR-T1D study with T1D underwent MRI hemodynamic evaluation. Inclusion criteria included age 12 to 21 years, T1D duration > 1 year (based on a clinical course consistent with T1D, at least one positive diabetes-associated antibody and rapid conversion to insulin requirement after diagnosis), and HbA1c ≤ 12%. Exclusion criteria were any comorbid conditions including heart failure, active or end stage liver disease, kidney disease (except microalbuminuria), inadequately treated thyroid disease, or rheumatologic disease; tobacco or marijuana use, pregnancy or breast feeding, medications altering the insulin sensitivity (diabetes medications other than insulin, oral steroid use, neuroleptics), ergot-related medications or triptan medications for migraine, diagnosis or history of psychosis, and diabetes of other causes. Full description of the BCQR-T1D study trial, power analysis, study safety monitoring, are described in the initial BCQR-T1D report¹⁷.

Baseline evaluations involved a fasting comprehensive metabolic panel along with anthropometric measurements including height and weight to calculate BMI z-score for sex and age by the least mean squares method, using 2000 Centers for Disease Control and Prevention growth charts. Body composition by dual-energy x-ray absorptiometry (Hologic, Marlborough, Massachusetts) was performed to determine fat mass and lean mass as previously reported²¹.

Intervention

Following the initial screening and baseline evaluation, participants were randomized to order of treatment phase per a cross-over assignment intervention model (placebo followed by BCQR or BCQR followed by placebo) (Figure 1). 1:1 randomization was conducted by a research pharmacist in a double-blinded fashion. 4 weeks of BCQR - (Cycloset™) using an 0.8 mg dose or identical placebo tablets (distributed by Belmar Compounding Pharmacy, Lakewood, CO) were provided for phase 1. This was followed by a 4 week washout period and then a 4 week phase 2 with the alternate treatment. Investigational medication in each phase was titrated as follows: BCQR (0.8 mg) or placebo 1 pill every morning for one week, then 2 pills every morning for one week, and finally 4 pills every morning for the remainder of the 4 weeks. If an uptitration step was not tolerated, participants were returned to the last tolerated dose for a few days and uptitration was attempted again. If again not tolerated, participants remained at the last tolerated dose as long as it was at least 1.6 mg daily. If unable to reach 1.6 mg daily, participants were withdrawn from the study. The Children’s Hospital Colorado research pharmacist maintained the blinding code and assigned participants to interventions to preserve blinding for the investigators and participants.

Figure 1.

CONSORT (Consolidated Standards of Reporting Trials) diagram of cross-over clinical trial in adolescents with type 1 diabetes mellitus.

Upon the completion of each phase of the intervention, participants presented to the local clinical and translational research center having taken a last dose of study drug that morning and underwent cardiovascular MRI, pressure hemodynamics, and peripheral vascular evaluation (described below).

Cardiovascular MRI

All participants underwent a cardiovascular MRI as described previously²². Briefly, phase-contrast MRI ECG-gated sequences were used to generate tissue magnitude and phase velocity maps of the aorta using a 1.5 or 3.0 Tesla system (Ingenia; Philips Medical Systems, Best, the Netherlands) with a phased-array body surface coil. Typical free breathing phase contrast-MRI sequence parameters were: TR: 5.8 ms, TE: 2.2–3.5 ms, resulting in temporal resolution of 14–28 milliseconds/40–50 cardiac phases, matrix: 160 256, flip angle: 25°, with 100% k-space sampling and no temporal interpolation, the cross-sectional pixel resolution ranged between 0.82×0.82 mm² and 1.56×1.56 mm² with a slice thickness of 8 mm. The phase-contrast-MRI acquisition time for the aortic plane varied between 2 and 3 minutes depending on heart rate. Given the different intrinsic differences in the aortic tissue composition as well as varying geometry along the thoracic aorta, central aortic stiffness parameters were measured separately in the ascending aorta (at the level of the right pulmonary artery), mid-descending aorta, and thoracoabdominal aorta. Velocity encoding values were selected to avoid an aliasing artifact and ranged between 100–200 cm/second.

Aortic Stiffness Analysis

The temporal segmentation of region-specific aortic contours was performed using the Circle CVI42 (version 5.9.3, Calgary, Canada). Aortic stiffness was calculated from vessel area changes and pressure measurements (obtained immediately prior to the MRI acquisition). Given that peripherally collected brachial blood pressure measurements overestimate the central aortic pressure values, we adjusted the peripheral measurements using an exponential pressure-area model as shown previously²³. Mean arterial pressure (MAP) was calculated as: MAP = diastolic blood pressure (DBP) + 0.4*PP. Central arterial stiffness was evaluated via pulse wave velocity (PWV). Non-invasively, PWV was calculated by using the Bramwell-Hill equation:

$$PWV={\left(\frac{A_{min}PP}{\rho (A_{max}-A_{min})}\right)}^{0.5}$$

where A_max and A_min represent the maximum and minimum aortic luminal areas, respectively, throughout the cardiac cycle, PP represents pulse pressure, and ρ is a blood density considered to have a constant value of 1060 kg/m³. We further calculated the local relative area change (RAC) defined as the difference between maximum and minimum areas divided by maximum value [(A_max−A_min)/A_max]×100%. Lastly, aortic distensibility was calculated as the ratio of RAC and pulse pressure.

Reactive hyperemia-peripheral arterial tonometry (RH-PAT)

Digital vascular response was evaluated with RH-PAT using the EndoPAT device (Itamar Medical Ltd., Caesarea, Israel). This technique combines flow-mediated dilatation with pneumatic fingertip probes with measures of arterial pulse wave amplitude, and provides a surrogate measure of digital endothelial function termed the reactive hyperemic index (RHI)^24,25. We further sampled logarithmically transformed RHI (Ln RHI) pulse wave signal with a matched cut-off value providing a similar index with better double-sided distribution. Lastly, a Framingham RHI was calculated from the Ln RHI waveform within 90–120 seconds post occlusion²⁶.

Peripheral Vascular Stiffness

Peripheral vascular stiffness was estimated by brachial artery distensibility (BrachD) and evaluated with the DynaPulse Pathway device (PulseMetric, Inc., San Diego California) as in our previous studies⁵. Dynapulse derives BrachD using pulse waveform analysis of arterial pressure signals obtained from the sphygmomanometer. Brachial artery waveforms were uploaded to an online analysis site and analyzed for measures of vascular stiffness. Brachial artery stiffness indices included distensibility estimated from waveform as dV/dP/V, and brachial artery compliance estimated as dV/dP.

Statistical Analysis

Analyses were performed with Prism (version 7.0 or higher; GraphPad Software Inc., La Jolla, CA) and SAS (version 9.4 or higher; SAS Institute, Cary, NC). Variables were checked for the distributional assumption of normality with normal plots, in addition to D’Agostino-Pearson, Shapiro-Wilk, and Kolmogorov-Smirnov tests. Baseline demographic and clinical characteristics of T1D participants were reported as mean or median values with corresponding SD or interquartile range, respectively, as dictated by the data distribution.

Prior to the evaluation of the effect of BCQR versus placebo therapy, interactions between treatment group and phase (order of BCQR versus placebo) in the cross-over design were tested in a mixed model. We observed no significant interactions (p<0.10) between treatment group and phase for any of the cardiovascular outcome variables. Consequently, adjusting for the randomization assignment of treatment order was not indicated.

The effect of the BCQR therapy was then evaluated using the 2-tailed paired t-test for normally distributed variables and paired Wilcoxon-rank test for non-uniformly distributed datasets. Central aortic stiffness measurements are directly derived or partly dependent on pressure-based loading conditions and therefore changes in all MRI measurements were also aadjusted for change in mean arterial pressure (MAP) using a linear mixed model. Comparative analysis between BCQR and placebo was graphically visualized using Gardner-Altman plots (Estimation Stats)²⁷. Significance for all analyses was set at a p-value of <0.05.

RESULTS

Clinical and demographic characteristics of T1D participants are reported in Table 1. A CONSORT (Consolidated Standards of Reporting Trials) diagram depicting the randomization and treatment phases is depicted in Figure 1. During phase one, 3 participants in the BCQR group and 5 participants in the placebo group did not participate in the MRI evaluation. Further, a single MRI acquisition at the level of ascending aorta, mid-descending aorta, and thoracoabdominal aorta was discarded in the BCQR group due to poor quality or noise, and a single MRI acquisition of the ascending aorta was removed from the placebo group due to acquisition plane mis-alignment. During phase 2, one MRI acquisition of the descending aorta was removed in the BCQR group due to poor quality. This yielded a final pair-wise comparison of 32 participants for the ascending aorta, 32 for the descending aorta, and 33 for the abdominal aorta.

Table 1.

Baseline characteristics of T1D Participants

	T1D (n = 34)
Age (years)	15.9 ± 2.6
Sex (%male)	13 (38.2%)
Diabetes duration (years)	5.8 (3.5 – 9.5)
HbA1c (%)	8.6 ± 1.1
BMI (kg/m2)	24.6 ± 5.8
BMI Percentile	71.4 ± 26.1
Lean body mass (kg)	43.8 ± 9.4
Fat free mass (kg)	45.8 ± 9.8
Body fat mass percentage (%)	32.7 ± 8.1
Lean fat mass percentage (%)	64 ± 8

Data reported as mean ± SD or median with corresponding IQR.

Pressure Hemodynamics Outcomes

Comparison of the pressure hemodynamic values between placebo and BCQR are reported in Table 2 and graphically represented in Figure 2. Compared to placebo, systolic blood pressure was significantly decreased with BCQR (Δ = −5 [95.0% CI: −7, −3] mmHg, p < 0.001) as was diastolic blood pressure (Δ = −2 [95.0% CI: −4.0, 0] mmHg, p = 0.039), pulse pressure (Δ = −3 [95.0% CI: −5, −1] mmHg, p = 0.015) and mean arterial pressure (Δ = −3 [95.0% CI: −5, −1] mmHg, p = 0.003).

Table 2.

Arterial Hemodynamics of Participants with T1D

	Placebo	Bromocriptine	P-value
Pressure Hemodynamics
Systolic blood pressure (mm Hg)	118 ± 9	113 ± 11	<0.001
Diastolic blood pressure (mm Hg)	69 ± 6	66 ± 7	0.039
Pulse pressure (mm Hg)	49 ± 7	46 ± 8	0.015
Mean arterial pressure (mm Hg)	88 ± 7	85 ± 8	0.002
Ascending Aorta
Pulse Wave Velocity (m/s)	4.0 ± 0.8	3.7 ± 0.6	0.018
Relative Area Change (%)	28.1 ± 6.8	30.7 ± 5.7	0.083
Distensibility (%/mm Hg)	0.59 ± 0.17	0.68 ± 0.17	0.017
Descending Aorta
Pulse Wave Velocity (m/s)	4.2 ± 0.6	4.1 ± 0.8	0.278
Relative Area Change (%)	25.9 ± 4.5	26.2 ± 5.7	0.779
Distensibility (%/mm Hg)	0.55 ± 0.12	0.59 ± 0.16	0.079
Abdominal Aorta
Pulse Wave Velocity (m/s)	4.2 ± 0.6	3.9 ± 0.4	0.007
Relative Area Change (%)	26.1 ± 3.8	27.3 ± 3.4	0.158
Distensibility (%/mm Hg)	0.56 ± 0.15	0.61 ± 0.11	0.013
Endothelial Function
Reactive Hyperemia Index	2.30 ± 0.56	1.99 ± 0.57	0.006
LN reactive hyperemia index	0.81 ± 0.23	0.65 ± 0.28	0.003
Framingham RHI Risk Score	0.87 (0.55 – 1.04)	0.75 (0.25 – 0.99)	0.010
Peripheral Vascular Stiffness
Brachial compliance (mL/mmHg)	0.059 ± 0.012	0.061 ± 0.013	0.156
Brachial distensibility (%/mmHg)	6.06 ± 1.16	6.25 ± 1.15	0.303

Data reported as mean ± SD or as median with corresponding IQR. LN = natural logarithm.

Figure 2.

Bromocriptine quick release (BCQR) improves overall pressure hemodynamics. BCQR reduced systolic A) and diastolic B) blood pressure as well as pulse pressure C) and mean arterial pressure D).

Aortic Stiffness Outcomes

Central aortic stiffness indices between placebo and BCQR are reported in Table 2 and graphically represented in Figure 3. In the ascending aorta, compared to the placebo group, BCQR significantly decreased PWV (Δ = −0.4 [95.0% CI: −0.1, −0.7] m/s, p = 0.018) and not significantly increased RAC (Δ = 2.6 [95.0% CI: 0.3, 4.6,] %, p = 0.158), and increased distensibility (Δ = 0.08 [95.0% CI: 0.15, 0.01] %/mmHg, p = 0.017). Changes measured in the ascending aorta remained significant upon adjustment for change in MAP, except for RAC. In the descending aorta, there were no significant differences between treatment groups for PWV (Δ = −0.1 [95.0% CI: 0.1, −0.3] m/s, p = 0.278), RAC (Δ = 0.3 [95.0% CI: 2.2, −1.5] %, p = 0.779), or distensibility (Δ = 0.04 [95.0% CI: 0.09, −0.01] %/mmHg, p = 0.079)). In the thoraco-abdominal aorta, BCQR significantly decreased PWV (Δ = −0.2 [95.0% CI: −0.1, −0.4] m/s, p = 0.007), and trended toward increased RAC (Δ = 1.3 [95.0% CI: 2.7, −0.2] %, p = 0.158) and significantly increased distensibility (Δ = 0.05 [95.0% CI: 0.09, 0.01] %/mmHg, p = 0.013). Reduction in PWV and distensibility measured in the thoraco-abdominal aorta remained significant after adjusting for change in MAP.

Figure 3.

Bromocriptine quick release (BCQR) improves central aortic stiffness in youth with T1D. Ascending aortic stiffness (top row) was improved as assessed by decreased pulse wave velocity, increased relative area change (approaching statistical significance), and increased distensibility. There were no differences in the descending aorta (middle row). Thoraco-abdominal stiffness was improved as evidenced by decreased pulse wave velocity and increased distensibility. All presented P-values are adjusted for change in mean arterial pressure.

Peripheral Vascular Stiffness Outcomes

Peripheral vascular stiffness measures from Dynapulse between placebo and BCQR are reported in Table 2 and graphically represented in Supplementary Figure S1. Compared to the placebo group, the BCQR group had a trend toward increased BrachD in the overall group (Δ = 0.26 [95.0% CI: −0.09, 0.63] %/mmHg, p= 0.156) but no significant difference in brachial artery compliance (Δ = 0.002 [95.0% CI: −0.001, 0.005] mL/mmHg, p= 0.303). However, stratification of BrachD by BMI category revealed that in the normal-weight BMI group, BrachD was significantly higher (improved) on BCQR versus placebo (6.95 ± 0.21 versus 6.42 ± 0.21 %/mmHg, p=0.015).

RH-PAT Outcomes

RH-EndoPAT measurements between placebo and BCQR are represented in Table 2 and graphically in Figure 4. Compared to the placebo group, participants on BCQR had decreased RHI (Δ = −0.34 [95.0% CI: −0.56, −0.10], P = 0.006), decreased Ln RHI (Δ = −0.17, [95.0% CI: −0.27, −0.06], p = 0.003), and reduced Framingham RHI (Δ = −0.18 [95.0% CI: −0.32, −0.04], p = 0.013).

Figure 4.

Bromocriptine therapy (BCQR) reduced endothelial response. This was demonstrated by - decreased reactive hyperemia index (RHI) A), - Ln RHI B) and by reduced Framingham RHI risk score C).

BMI Stratification Analysis

In order to explore the effect of BMI on pressure hemodynamics and vascular biomechanics, we performed sub-analysis comparing the effect of BCQR therapy separately in normal-weight participants (n=18) and in overweight/obese participants (n=16). Subgroup analysis focusing on blood pressure is depicted in Supplementary Figure S2. Systolic blood pressure was uniformly improved in both normal-weight (Δ = −5 [95.0%CI: −7, −1] mmHg, p= 0.009) and overweight/obese participants (Δ = −5 [95.0%CI: −8, −2] mmHg, p= 0.003). There was no significant change in diastolic blood pressure in normal weight participants (Δ = 0 [95.0%CI: −3, 2] mmHg, p= 0.767) but there was a reduction in diastolic blood pressure in overweight/obese subgroup (Δ = −5 [95.0%CI: −7, −1] mmHg, p= 0.012).

BMI subgroup analysis focusing on the ascending aortic stiffness is depicted in Supplementary Figure S3. PWV was significantly improved in normal-weight participants (Δ = −0.4 [95.0%CI: −0.7, 0] m/s, p= 0.030) and trended toward improvement but lacked statistical significance in overweight/obese participants (Δ = −0.4 [95.0%CI: −0.8, 0.1] m/s, p= 0.072). There was no significant change in RAC in normal weight participants (Δ = 2.1 [95.0%CI: −1.0, 5.3] %, p= 0.229). C) but there was a significant increase RAC in overweight/obese subgroup (Δ = 3.2 [95.0%CI: 0.9, 5.9] %, p= 0.035). There was a trend toward increased distensibility in both normal-weight (Δ = 0.09 [95.0%CI: −0.01, 0.19] %/mmHg, p= 0.084) and overweight/subgroups (Δ = 0.07 [95.0%CI: 0.01, 0.14] %/mmHg, p= 0.068) but without statistical significance.

Lastly, a BMI subgroup analysis focusing on the peripheral BrachD is graphically depicted in Supplementary Figure S4. There was a significant increase in BrachD in normal-weight participants (Δ = 0.56 [95.0%CI: 0.10, 0.98] %/mmHg, p= 0.023) but no observed effect in overweight/obese participants (Δ = 0.25 [95.0%CI: −0.09, 0.63] %/mmHg, p= 0.156).

DISCUSSION

The main cardiovascular findings of this placebo-controlled, random-order, double-blinded, cross-over study were that in T1D BCQR improves: 1) standard blood pressure hemodynamics, 2) multiple measures of central aortic stiffness in the ascending and thoracoabdominal aorta as evidenced by MRI, 3) peripheral arterial stiffness in normal weight youth as evidenced by Dynapulse derived brachial artery distensibility, with a trend towards improvement in peripheral vascular stiffness in the overall group. This is the first study demonstrating the effect of BCQR versus placebo therapy on arterial health in a T1D population, and the first BCQR data in a pediatric population with diabetes of any kind. We comprehensively demonstrate the effect of BCQR on large elastic and medium size muscular arteries.

Increased blood pressure and aortic stiffness are well described in people with T1D and T2D. Hypertension further accelerates end-organ damage and is significantly associated with additional cardiovascular risk factors including coronary, cerebrovascular, and kidney diseases²⁸. The primary consequence of large arterial or aortic stiffening is a loss of compliance and buffering potential, resulting in an increase in antegrade and retrograde pulse wave propagation. Consequently, the pulsatile load generated by the left ventricle is augmented, which may lead to increased stress on large arteries, peripheral arterial vasculopathy, and possibly inflict end-organ damage^18,29. Both increased blood pressure and aortic stiffness significantly increase systemic ventricular afterload and may have a detrimental long-term effect on overall cardiac function, leading to ventricular hypertrophy as we previously detected in T1D³⁰. Tonometrically measured stiffness by PWV has been previously shown to be increased in youth and young adults with T1D^31,32. Applanation tonometry samples signals collected in primary or secondary aortic branches which yields measurements influenced by non-uniform vessel behavior and wave reflections due to impedance mismatch. Furthermore, the anatomic intraluminal distance between two measurement sites is typically estimated and is not reflective of the real anatomic-physiologic course of blood flow. Consequently, tonometry provides a global measure of large artery stiffness. In comparison to applanation tonometry, MRI can evaluate central arterial stiffness directly at a specific aortic site and provides measurements which are more respective of local flow hemodynamic conditions ³³. This advantage of MRI allows for sensitive detection of change, such as that seen in our EMERALD trial in which MRI was able to detect improvement in central aortic stiffness in youth with T1D following 3 months of treatment with metformin vs. placebo⁵.

In the BCQR-T1D study, we now show that BCQR therapy improves central aortic stiffness in the ascending and thoracoabdominal aorta. The most prominent change was detected in the ascending aortic stiffness indices, where all considered MRI markers demonstrated significant improvement. These findings are consistent with the EMERALD trial in which we also demonstrated improvement in the metformin treated group in the ascending aorta, but no effect in the descending aorta⁵. One explanation for these regional differences in responses may be different aortic composition in each region and/or variable endothelial cell responsiveness to circulating exogenous factors and shear forces in different aortic regions^34,35, highlighting the importance of evaluating segment-specific aortic stiffness. Additionally, we observed improvement in the thoraco-abdominal stiffness with BCQR, as evidenced by both locally decreased PWV and increased distensibility. This finding again suggests that different aortic segments might respond differently to BCQR, which targets the sympathetic branch of the autonomic nervous system which is unevenly distributed via the thoracic aortic plexus along the thoracoabdominal aorta³⁶.

Brachial artery stiffness and distensibility demonstrated a trend towards improvement in the overall group, along with significant improvements when just the normal-weight group was evaluated, supporting the findings of the MRI-derived central aortic stiffness measures. Future studies with a larger sample size and longer treatment duration are now required, along with further studies employing high-resolution imaging of the peripheral vasculature capable of accurate measurements of baseline peripheral arterial size to fully understand the effect on medium sized muscular arteries.

The results from the BCQR-T1D study further also suggest that BCQR therapy reduces digital endothelial function as estimated by RH-PAT-assessed RHI and its associated indices. Reduction in RHI is typically perceived as impaired endothelial signaling mediating the local autoregulation in response to transient tissue hypoxia and accumulation of vasodilator metabolites^26,37. In the presence of an exogenous sympatholytic mediator such as bromocriptine, the constant sympathetic tone to the vasculature is decreased, leading to a maintained state of vasodilation and a systemic hypotensive state^38,39. This effect was demonstrated in experimental work by Schobel et al showing that bromocriptine administration acutely increases forearm blood flow in healthy individuals, while simultaneously decreasing sympathetic vascular tonicity⁴⁰. This state has been shown to result in impaired vascular responsiveness, as demonstrated by absent normalization in response to a cold-stressor test, or by administration of nitroprusside³⁷. We speculate that a similar effect might be responsible for the RHI results observed in the BCQR-T1D study. Unfortunately to our best knowledge, there are no prior clinical studies investigating the responsiveness of RHI to bromocriptine therapy in other populations, and therefore we do not yet have comparative results. Further longer-term studies are now needed to determine the short-term vs. longer term vascular impacts of BCQR in T1D.

It is thought that BCQR primarily mediates vascular tone by central inhibition of the sympathetic drive, limiting overactivation of the hypothalamic-pituitary-adrenal axis¹⁵. Results from our current study cannot delineate whether observed attenuation in central aortic stiffness is mediated centrally, or whether local vascular wall remodeling occurred, or a combination of the two. Prior studies indicate that both central and peripheral mechanisms inhibiting norepinephrine release might simultaneously occur⁴⁰. Additionally, bromocriptine has been shown to inhibit proliferation of vascular smooth muscle cells, arguing for a remodeling effect as well as a central effect⁴¹.

We would like to acknowledge several limitations pertinent to the BCQR-T1D study. First, the 4-week intervention period might have not been sufficiently long to fully appreciate the effect of the BCQR on central and peripheral vascular stiffness. Second, the peripherally sampled pressure hemodynamic measurements were used uniformly for the calculation of the aortic distensibility. True pressure waveforms are clinically obtainable only via invasive catheterization which is not feasible in this pediatric population. Lastly, longer term studies and studies of other age groups are required to determine the cardiovascular effects of treatment with BCQR in T1D. However, in our larger BCQR-T1D study, we did find improved blood pressure in adults with T1D as well, suggesting similar effects in adults, but we were unable to perform MRI scans on the adults in the BCQR-T1D study due to financial constraints.

Conclusions

Elevated central aortic stiffness, blood pressure, and likely peripheral stiffness is present in youth with T1D, and attenuated by 4-weeks of BCQR therapy. Arterial stiffness and blood pressure are predictors of all cause cardiovascular events. As such, BCQR therapy might serve as a clinical intervention to reduce cardiovascular risk in youth with T1D, supported by evidence of reduced major adverse cardiovascular events in adults with T2D adults after just 1-year of treatment with BCQR^15,16. Additionally, BCQR improves standard pressure hemodynamics, likely to improve long-term cardiovascular outcomes. However, longer-term study and study of additional age groups is needed to determine the significance of the observed reduction in RHI and BCQR’s ultimate impact on cardiovascular morbidity and mortality in T1D.

PERSPECTIVES

There is a strong association between an earlier age of T1D diagnosis and adverse cardiovascular outcomes. Bromocriptine quick release (BCQR) formulation is approved for therapeutic use in adults with type 2 diabetes with documented risk lowering effect on adverse cardiovascular events. However, the effect of BCQR on cardiovascular physiologic parameters has never been directly investigated. In this placebo-controlled, random-order, double-blinded, cross-over study we observed that BCQR improve blood pressure hemodynamics and attenuates central aortic stiffness in youth with T1D. Given the strong prognostic value of both arterial stiffness and blood pressure hemodynamics, BCQR might represent a new therapeutic avenue in this patient population. Presented results warrant and further validation in larger patient cohort over longer time period.

PATHOPHYSIOLOGICAL NOVELTY AND RELEVANCE.

WHAT IS NEW?

This is the first study demonstrating the effect of BCQR versus placebo therapy on arterial health in a T1D population, and the first BCQR data in a pediatric population with diabetes of any kind. We comprehensively demonstrate the effect of BCQR on large elastic and medium size muscular arteries

WHAT IS RELEVANT?

Increased blood pressure and aortic stiffness are well described in people with T1D and T2D. The development of therapeutic strategies to slow progression of systemic vascular disease in youth with T1D is of critical importance.

CLINICAL/PATHOPHYSIOLOGICAL IMPLICATIONS

BCQR improves in youth with T1D 1) standard blood pressure hemodynamics, 2) central aortic stiffness in the ascending and thoracoabdominal aorta as evidenced by MRI, 3) peripheral arterial stiffness in normal weight youth as evidenced by improved brachial artery distensibility.

ABBREVIATIONS

BCQR

bromocriptine quick release

BrachD

brachial distensibility

cIMT

carotid intima-media thickness

MAP

mean arterial pressure

PWV

pulse wave velocity

RAC

relative area change

RHI

reactive hyperemia index

T1D

type 1 diabetes