Characterization of regional left ventricular function in nonhuman primates using magnetic resonance imaging biomarkers: a test-retest repeatability and inter-subject variability study

Comparison of cine and real-time cardiac MRI in rhesus macaques

Cardiac MRI in rhesus macaques, a species of major relevance for preclinical studies on biological therapies, requires artificial ventilation to realize breath holding. To overcome this limitation of standard cine MRI, the feasibility of Real-Time (RT) cardiac MRI has been tested in a cohort of ten adult rhesus macaques using a clinical MR-system. In spite of lower tissue contrast and sharpness of RT-MRI, cardiac functions were similarly well assessed by RT-MRI compared to cine MRI (similar intra-subject repeatability). However, systematic underestimation of the end-diastolic volume (31 ± 9%), end-systolic volume (20 ± 11%), stroke volume (40 ± 12%) and ejection fraction (13 ± 9%) hamper the comparability of RT-MRI results with those of other cardiac MRI methods. Yet, the underestimations were very consistent (< 5% variability) for repetitive measurements, making RT-MRI an appropriate alternative to cine MRI for longitudinal studies. In addition, RT-MRI enabled the analysis of cardio-respiratory coupling. All functional parameters showed lower values during expiration compared to inspiration, most likely due to the pressure-controlled artificial ventilation. In conclusion, despite systematic underestimation of the functional parameters, RT-MRI allowed the assessment of left ventricular function in macaques with significantly less experimental effort, measurement time, risk and burden for the animals compared to cine MRI.

A Role of Glucose Overload in Diabetic Cardiomyopathy in Nonhuman Primates

Type 2 diabetes (T2D) plays a major role in the development of heart failure. Patients with T2D have an increased risk to develop HF than healthy subjects, and they always have very poor outcomes and survival rates. However, the underlying mechanisms for this are still unclear. To help develop new therapeutic interventions, well-characterized animal models for preclinical and translational investigations in T2D and HF are urgently needed. Although studies in rodents are more often used, the research findings in rodents have often failed to be translated into humans due to the significant metabolic differences between rodents and humans. Nonhuman primates (NHPs) serve as valuable translational models between basic studies in rodent models and clinical studies in humans. NHPs can recapitulate the natural progress of these diseases in humans and study the underlying mechanism due to their genetic similarity and comparable spontaneous T2D rates to humans. In this review, we discuss the importance of using NHPs models in understanding diabetic cardiomyopathy (DCM) in humans with aspects of correlations between hyperglycemia and cardiac dysfunction progression, glucose overload, and altered glucose metabolism promoting cardiac oxidative stress and mitochondria dysfunction, glucose, and its effect on cardiac resynchronization therapy with defibrillator (CRT-d), the currently available diabetic NHPs models and the limitations involved in the use of NHP models.

Cardiac MRI in common marmosets revealing age-dependency of cardiac function

The aim of this study was to establish a feasible and robust magnetic resonance imaging protocol for the quantitative assessment of cardiac function in marmosets and to present normal values of cardiac function across different ages from young adult, middle-aged, to very old clinically healthy animals. Cardiac MRI of 33 anesthetized marmosets at the age of 2-15 years was performed at 9.4 T using IntraGate-FLASH that operates without any ECG-triggering and breath holding. Normalized to post-mortem heart weight, the left ventricular end-diastolic volume (LV-EDV) was significantly reduced in older marmosets. The LV end-systolic volume (LV-ESV) and the LV stroke volume (LV-SV) showed a similar trend while the LV ejection fraction (LV-EF) and wall thickening remained unchanged. Similar observations were made for the right ventricle. Moreover, the total ventricular myocardial volume was lower in older monkeys while no significant difference in heart weight was found. In conclusion, IntraGate-FLASH allowed for quantification of left ventricular cardiac function but seems to underestimate the volumes of the right ventricle. Although less strong and without significant sex differences, the observed age related changes were similar to previously reported findings in humans supporting marmosets as a model system for age related cardiovascular human diseases.

2D/3D CMR tissue tracking versus CMR tagging in the assessment of spontaneous T2DM rhesus monkeys with isolated diastolic dysfunction

BACKGROUND

Spontaneous T2DM in rhesus monkeys manifests as isolated diastolic dysfunction in the early stage of diabetic cardiomyopathy, similar to humans. Myocardial deformation measurements have emerged as a superior way to measure left ventricular (LV) function in the early stage of cardiac dysfunction, making it possible to further evaluate early-stage LV dysfunction in spontaneous T2DM rhesus monkeys.

METHODS

Spontaneous T2DM rhesus monkeys with isolated diastolic dysfunction (T2DM-DD, n = 10) and corresponding nondiabetic healthy animals (ND, n = 9) were prospectively scanned for a CMR study. Circumferential and longitudinal peak systolic strain (Ecc, Ell), time to peak strain (tEcc, tEll) and peak diastolic strain rate (CSR, LSR) obtained from 2D/3D CMR-TT were compared with those obtained from CMR tagging separately. In addition, all CMR imaging protocols were performed twice in 9 ND animals to assess test-retest reproducibility.

RESULTS

Compared with the ND group, the T2DM-DD monkeys demonstrated significantly impaired LV Ecc (- 10.63 ± 3.23 vs - 14.18 ± 3.19, p < 0.05), CSR (65.50 ± 14.48 vs 65.50 ± 14.48, p < 0.01), Ell (- 9.11 ± 2.59 vs - 14.17 ± 1.68, p < 0.05), and LSR (59.43 ± 19.17 vs 108.46 ± 22.33, p < 0.01) with the tagging. Only Ecc (- 13.10 ± 2.47 vs - 19.03 ± 3.69, p < 0.01) and CSR (148.90 ± 31.27 vs 202.00 ± 51.88, p < 0.01) were significantly reduced with 2D CMR-TT, and only Ecc (- 13.77 ± 1.98 vs - 17.26 ± 3.78, p < 0.05) was significantly reduced with 3D CMR-TT. Moreover, 2D/3D CMR-TT-derived Ecc and CSR correlated with the corresponding tagging values collectively, with a statistically significant ICC value (p < 0.05). Test-retest repeatability analysis showed that most tagging-derived biomarkers had acceptable repeatability (p < 0.01). In addition, 2D CMR-TT-derived indicators were poorer than those derived from the tagging method but better than those obtained using the 3D method, with larger ICCs except for tEcc (p < 0.05).

CONCLUSIONS

LV systolic and diastolic deformations were impaired in spontaneous T2DM rhesus monkeys previously diagnosed with isolated diastolic dysfunction by echocardiography. The 2D CMR-TT-derived Ecc and CSR were effective in the evaluation of the myocardial systolic and diastolic functions of early-diabetic cardiomyopathy, with relatively higher test-retest reproducibility and acceptable correlation with the tagging method compared with the 3D CMR-TT method.