Mapping the Unseen: In Vivo CEST-MRI of Creatine Reveals Improved Cardiac Energetics in Subjects with Obesity Following Bariatric Surgery

Abstract

Background

Obesity is associated with derangement of cardiac metabolism and the development of subclinical cardiovascular disease. This prospective study examined the impact of bariatric surgery on cardiac function and metabolism.

Methods

Subjects with obesity underwent cardiac magnetic resonance imaging (CMR) at Massachusetts General Hospital before and after bariatric surgery between 2019 and 2021. The imaging protocol included Cine for global cardiac function assessment and creatine chemical exchange saturation transfer (CEST) CMR for myocardial creatine mapping.

Results

Thirteen subjects were enrolled, and 6 subjects [mean BMI 40.5 ± 2.6] had completed the second CMR (i.e. post-surgery), with a median follow-up of 10 months. The median age was 46.5 years, 67% were female, and 16.67% had diabetes. Bariatric surgery led to significant weight loss, with achieved mean BMI of 31.0 ± 2.0. Additionally, bariatric surgery resulted in significant reduction in left ventricular (LV) mass, LV mass index, and epicardial adipose tissue (EAT) volume. This was accompanied by slight improvement in LV ejection fraction compared to baseline. Following bariatric surgery, there was a significant increase in creatine CEST contrast. Subjects with obesity had significantly lower CEST contrast compared to subjects with normal BMI (n = 10), but this contrast was normalized after the surgery, and statistically similar to non-obese cohort, indicating an improvement in myocardial energetics.

Conclusions

CEST-CMR has the ability to identify and characterize myocardial metabolism in vivo non-invasively. These results demonstrate that in addition to reducing BMI, bariatric surgery may favorably affect cardiac function and metabolism.

Introduction

The heart is constantly consuming energy throughout its lifetime, with energetic reserve capacity to respond to increased demand. Cardiac metabolic derangements has previously been reported in several clinical contexts including myocardial infarction, heart failure, and obesity [1, 2]. While cardiac ATP at rest is often maintained in obesity, previous study showed impaired cardiac energy reserve via creatine kinase reaction and it may be involved in the progression to heart failure [1, 2]. Efforts to study cardiac energy substrate metabolism in vivo has been a field of active research for more than a century [3]. Among others, chemical exchange saturation transfer (CEST) is a promising metabolic imaging technique which can be used for creatine mapping in the heart [4]. This prospective study examined the impact of bariatric surgery on cardiac function and metabolism.

Methods

Subjects with moderate to severe obesity (body mass index, BMI ≥ 35 kg/m²) who underwent gastric sleeve surgery and subjects with normal BMI (n = 10) were consented and recruited with approval from the Institutional Review Board of the Massachusetts General Hospital and conformed the Declaration of Helsinki. Cardiac magnetic resonance imaging (CMR) was performed before and after surgery on a 3 T MRI Skyra clinical scanner (Siemens Medical, Germany). The imaging protocol included conventional CINE, fat–water imaging, native T1/T2 relaxometry, and CEST-CMR for myocardial creatine mapping. In addition, a phantom doped with varying creatine concentrations was built and scanned to validate the relationship between asymmetric magnetization transfer ratio (MTR_asym) and creatine concentration. Creatine CEST-CMR scans were performed using a previously published CEST preparation pulse train [4] coupled with a reduced field of view echo planar imaging readout to reduce MT effects and was gated with ECG and respiratory navigation. The segmentation of CINE images was performed on Segment software (Medviso, Sweden), while segmentation of visceral fat, native T1, and T2 was performed with Osirix (Pixmeo, Switzerland). Post-processing of CEST images was performed with custom-written software in Python. Data are presented as mean ± standard errors of the means. The results were compared using paired t-test. One-way ANOVA followed by Bonferroni test was used to compare differences of means among more than two groups. p-value < 0.05 was considered significant.

Results

Six subjects [mean BMI 40.5 ± 2.6 kg/m²] had completed the second CMR (i.e. post-surgery), with a median follow-up of 10 months. The median age was 46.5 years, 67% were female, and 16.67% had type 2 diabetes. Bariatric surgery induced significant reduction in body weight (Fig. 1A) with achieved mean BMI of 31.0 ± 2.0 kg/m²(Fig. 1B). In addition, bariatric surgery was also associated with significant reduction in left ventricular (LV) mass, LV mass index (i.e. LV mass Indexed to body surface area), and epicardial adipose tissue (EAT) volume (Fig. 1C–E). This was accompanied by an improvement in LV ejection fraction compared to baseline value (Fig. 1F). No significant changes were found in end-diastolic and systolic volume (129 ± 12 to 127 ± 7 ml and 58 ± 2 to 49 ± 4 ml, respectively), global longitudinal strain (17.9 ± 1.5% vs. 17.7 ± 0.9%), as well as myocardial native T1 and T2 values (Fig. 1G–H). Representative MRI images of Dixon-based water and fat content, CINE, and native T1 and T2 maps, respectively, are depicted in Fig. 1I–K. Phantom experiment revealed creatine concentration dependent MTR_asym(Fig. 1L). A significant increase in creatine CEST contrast was observed following bariatric surgery (Fig. 1M–N). Compared to normal-weight subjects (n = 10), CEST contrast was significantly lower in subjects with obesity; and it was normalized after the surgery and statistically similar to those with normal BMI, suggesting an improvement in myocardial energetics (Fig. 1O).

Fig. 1.

The impact of bariatric surgery on cardiac function and metabolism. Changes in body weight (A), body mass index (B), LV mass (C), LV mass index (D), epicardial fat (E), LV ejection fraction (F), T1 value (G) and T2 value (H) after bariatric surgery. (I-K) Representative images of fat–water content, cine CMR, and T1 and T2 maps, respectively. (L) Graphs shows creatine signal (MTR_asym) for each concentration obtained from CEST-CMR phantom experiment. (M) Representative masked images of cardiac CEST-CMR at 1.8 ppm. (N) Myocardial creatine content before and after bariatric surgery. (O) Comparison of CEST contrast between subjects with normal BMI and subjects with obesity. LV indicates left ventricle; CMR indicates cardiac magnetic resonance imaging. *p < 0.05, **p < 0.01, ***p < 0.001

Discussion

In line with previous studies [5, 6], our study demonstrates that bariatric surgery in addition to reducing body weight and BMI, could significantly reduce LV mass and EAT. However, our study also provides some new information, specifically the observation of an improvement in LV ejection fraction and myocardial creatine content detected by CEST-CMR following bariatric surgery. Although the mechanism of actions underlying LV mass regression and EAT reduction is unclear, weight loss in obesity is known to improve LV geometry and possible candidate pathways include neurohormonal pathways that might also improve CV outcomes. Furthermore, it may be speculated that, at least in part, LVEF improvement after surgery resulted from the reduction in EAT.

In concordance with our findings for myocardial energetics, a previous study reported metabolic derangement in cardiac energetics in obesity [2], specifically, the myocardial creatine signal measured by CEST-CMR, was lower in subjects with obesity compared to subjects with normal BMI [2]. A strong correlation between increased body and visceral fat volumes and reduced myocardial energetics was also observed [2]. Our findings extend this observation as we demonstrate improved cardiac structure and function as well as normalization of cardiac energetics after bariatric surgery. This study also highlights that bariatric surgery reverses the energetic changes and might be a potential modifiable pathway to prevent the progression to heart failure. Future studies should include a larger sample size with a separate appropriate control group. Furthermore, incorporating direct measurements of phosphocreatine/ATP using ³¹P-MRS and cardiac arterio-venous gradients into future studies would enable a more comprehensive assessment of cardiac metabolism and provide additional insights into the reliability of creatine mapping using CEST-MRI, as well as its correlation with cardiac energetics.

Conclusion

These results demonstrate that in addition to reducing BMI, bariatric surgery may favorably affect cardiac function and metabolism. An important aspect is the ability to measure myocardial energetics in vivo using conventional proton-based MRI scanners without invasive biopsy or ionizing radiation. This allows for repeated measurements in longitudinal studies and the potential for in vivo metabolic imaging to become a valuable clinical tool.

Key points.

• This study examines the impact of bariatric surgery on cardiac function and metabolism.

• Subjects with obesity has lower cardiac creatine levels.

• Bariatric surgery reduces weight, BMI, LV mass index, and epicardial fat.

• These changes are accompanied improvement in LVEF and cardiac energetics.

Data Availability

The data that support the findings of this study are available on request from the corresponding author.