Effects of Bariatric Surgery on Cardiovascular Function
Effects of Bariatric Surgery on Cardiovascular Function
This study was performed to assess the effects of bariatric surgery on cardiac ectopic fat accumulation in patients with type 2 diabetes, with a focus on pericardial fat, which has not been studied in detail before. We found a substantial reduction in ectopic cardiac fat, after 16-weeks follow-up. Interestingly, we observed a differential response of the pericardial fat layers after RYGB surgery, which has not been shown before. Pericardial fat consists of two layers, epicardial fat (located between the myocardium and visceral pericardium) and paracardial fat (located outside the parietal pericardium). Epicardial fat is a visceral fat depot, originating from mesothelial cells, and it is supplied by branches of the coronary arteries. Epicardial fat is thought to have several beneficial functions, for instance serving as a buffer to protect the myocardium from a toxic overload of free fatty acids (FFAs) and, on the other hand, supplying FFAs as an immediate energy source for the cardiac muscle in times of need. Furthermore, it might protect the coronary arteries from torsion. However, increased epicardial fat volume has been associated with insulin resistance, type 2 diabetes and CVD. Studies that assessed epicardial fat volume with MRI showed that epicardial fat volume is related to visceral fat volume at baseline and decreases after diet-induced weight loss and exercise in obese subjects. Previous studies with echocardiography revealed that epicardial fat thickness decreases after diet-induced weight loss.
Bariatric surgery also leads to a decrease in epicardial fat in obese subjects as measured by echocardiography and MRI. In our study, there was a decrease in epicardial fat volume as well; however, the relative decrease in paracardial fat volume was much higher. The role of paracardial fat is currently less clear. Some studies have shown that paracardial fat is a better predictor of cardiovascular risk than epicardial fat. Furthermore, a 6-month exercise intervention in patients with type 2 diabetes led to a decrease in visceral and paracardial, but not epicardial fat. This study and our findings might suggest that exercise and weight loss in patients with type 2 diabetes have differential effects on paracardial adipose tissue as opposed to epicardial adipose tissue. In contrast, a very recent study showed no differences in epicardial nor in paracardial fat after bariatric surgery. However, the follow-up duration in that study was shorter (mean 80 ± 24 days), as compared to our follow-up, which might explain the different results. Long-term follow-up of our patient group would be interesting to determine whether epicardial fat response is delayed compared with visceral and subcutaneous fat.
We also observed that the abdominal fat compartments did not respond equally to the surgically induced weight loss. The observed preferential loss of visceral fat as compared to subcutaneous fat confirms findings of earlier studies in obese subjects after bariatric surgery and in patients with type 2 diabetes after a VLCD.
We did not observe a decrease in myocardial TG after RYBG surgery. Gaborit et al. also did not show a decrease in myocardial TG content 6 months after bariatric surgery. It is an interesting finding that previous studies with dietary-induced weight loss interventions did find decreased myocardial TG in obese subjects with and without type 2 diabetes, whereas this is not found in studies on weight loss induced by bariatric surgery. It could be suggested that the surgical-induced weight loss has a differential effect on myocardial TG as compared to diet-induced weight loss. It is, however, unclear what the cause of this difference might be. It could be that a major surgery results in a higher inflammatory stress response as compared to dietary intervention. However, the C-reactive protein levels of our patients significantly decreased 4 months after surgery. A rapid decrease in markers of a chronic low grade inflammatory state after bariatric surgery has been found by others as well. In contrast to dietary interventions, bariatric surgery also directly induces significant changes in gastro-intestinal hormone secretion, which might contribute to the differences. The exact mechanism explaining the differences between dietary- and surgical-induced weight loss on myocardial TG content therefore warrants further investigation.
The high myocardial TG content at baseline in our study is striking. Other studies report myocardial TG contents of around 1·0, whereas we found a value of 1·18 ± 0·44. This high baseline value might indicate more severe cardiac lipotoxicity in our patient group and concomitant less flexibility of the myocardial TG content.
After the RYGB, we found a (nonsignificant) decrease in PWV, suggesting an improvement of central arterial stiffness. We did not observe an effect of RYGB surgery on cardiac function. The most important parameter for diastolic function, the E/A ratio, however, was already within the normal range (>1·0) at baseline, with a mean value of 1·26 ± 0·36. Therefore, an improvement in diastolic function might not be expected. In a previous study that did find improved diastolic function after weight loss using a VLCD, the E/A ratio at baseline was lower as compared to our study. In contrast, diastolic cardiac function did improve in the study by Gaborit et al. 6 months after bariatric surgery, even though a high E/A ratio was present at baseline as well. The difference might be explained by the fact that only 26% of the patients in that study had type 2 diabetes, whereas all of our patients had insulin-dependent type 2 diabetes. Furthermore, in our study, only two of the 10 patients were able to stop insulin treatment 4 months after the bypass surgery. It can therefore not be excluded that continued hyperinsulinemia may have influenced cardiac function.
A limitation of our study is the small sample size. Because of the narrow interior of the MRI scanner and the maximum allowable weight of the MRI table, many candidates for bariatric surgery are not eligible for MR studies. However, our study was successful in showing substantial changes of bariatric surgery on ectopic fat accumulation and in particular provided new data on fat distribution in epi- and paracardial fat. Another limitation is the relatively short follow-up period, which could be an explanation for the fact that we did not find statistically significant differences in cardiovascular function parameters between the two time points.
In conclusion, this study shows that weight loss induced by bariatric surgery leads to a decrease in abdominal and pericardial fat depots, with a higher relative decrease in visceral and paracardial fat volumes. These findings contribute to the existing evidence suggesting tissue-specific changes in body fat distribution after weight loss and exercise interventions. However, the decrease in pericardial fat did not lead to improved cardiovascular function after the RYGB.
Discussion
This study was performed to assess the effects of bariatric surgery on cardiac ectopic fat accumulation in patients with type 2 diabetes, with a focus on pericardial fat, which has not been studied in detail before. We found a substantial reduction in ectopic cardiac fat, after 16-weeks follow-up. Interestingly, we observed a differential response of the pericardial fat layers after RYGB surgery, which has not been shown before. Pericardial fat consists of two layers, epicardial fat (located between the myocardium and visceral pericardium) and paracardial fat (located outside the parietal pericardium). Epicardial fat is a visceral fat depot, originating from mesothelial cells, and it is supplied by branches of the coronary arteries. Epicardial fat is thought to have several beneficial functions, for instance serving as a buffer to protect the myocardium from a toxic overload of free fatty acids (FFAs) and, on the other hand, supplying FFAs as an immediate energy source for the cardiac muscle in times of need. Furthermore, it might protect the coronary arteries from torsion. However, increased epicardial fat volume has been associated with insulin resistance, type 2 diabetes and CVD. Studies that assessed epicardial fat volume with MRI showed that epicardial fat volume is related to visceral fat volume at baseline and decreases after diet-induced weight loss and exercise in obese subjects. Previous studies with echocardiography revealed that epicardial fat thickness decreases after diet-induced weight loss.
Bariatric surgery also leads to a decrease in epicardial fat in obese subjects as measured by echocardiography and MRI. In our study, there was a decrease in epicardial fat volume as well; however, the relative decrease in paracardial fat volume was much higher. The role of paracardial fat is currently less clear. Some studies have shown that paracardial fat is a better predictor of cardiovascular risk than epicardial fat. Furthermore, a 6-month exercise intervention in patients with type 2 diabetes led to a decrease in visceral and paracardial, but not epicardial fat. This study and our findings might suggest that exercise and weight loss in patients with type 2 diabetes have differential effects on paracardial adipose tissue as opposed to epicardial adipose tissue. In contrast, a very recent study showed no differences in epicardial nor in paracardial fat after bariatric surgery. However, the follow-up duration in that study was shorter (mean 80 ± 24 days), as compared to our follow-up, which might explain the different results. Long-term follow-up of our patient group would be interesting to determine whether epicardial fat response is delayed compared with visceral and subcutaneous fat.
We also observed that the abdominal fat compartments did not respond equally to the surgically induced weight loss. The observed preferential loss of visceral fat as compared to subcutaneous fat confirms findings of earlier studies in obese subjects after bariatric surgery and in patients with type 2 diabetes after a VLCD.
We did not observe a decrease in myocardial TG after RYBG surgery. Gaborit et al. also did not show a decrease in myocardial TG content 6 months after bariatric surgery. It is an interesting finding that previous studies with dietary-induced weight loss interventions did find decreased myocardial TG in obese subjects with and without type 2 diabetes, whereas this is not found in studies on weight loss induced by bariatric surgery. It could be suggested that the surgical-induced weight loss has a differential effect on myocardial TG as compared to diet-induced weight loss. It is, however, unclear what the cause of this difference might be. It could be that a major surgery results in a higher inflammatory stress response as compared to dietary intervention. However, the C-reactive protein levels of our patients significantly decreased 4 months after surgery. A rapid decrease in markers of a chronic low grade inflammatory state after bariatric surgery has been found by others as well. In contrast to dietary interventions, bariatric surgery also directly induces significant changes in gastro-intestinal hormone secretion, which might contribute to the differences. The exact mechanism explaining the differences between dietary- and surgical-induced weight loss on myocardial TG content therefore warrants further investigation.
The high myocardial TG content at baseline in our study is striking. Other studies report myocardial TG contents of around 1·0, whereas we found a value of 1·18 ± 0·44. This high baseline value might indicate more severe cardiac lipotoxicity in our patient group and concomitant less flexibility of the myocardial TG content.
After the RYGB, we found a (nonsignificant) decrease in PWV, suggesting an improvement of central arterial stiffness. We did not observe an effect of RYGB surgery on cardiac function. The most important parameter for diastolic function, the E/A ratio, however, was already within the normal range (>1·0) at baseline, with a mean value of 1·26 ± 0·36. Therefore, an improvement in diastolic function might not be expected. In a previous study that did find improved diastolic function after weight loss using a VLCD, the E/A ratio at baseline was lower as compared to our study. In contrast, diastolic cardiac function did improve in the study by Gaborit et al. 6 months after bariatric surgery, even though a high E/A ratio was present at baseline as well. The difference might be explained by the fact that only 26% of the patients in that study had type 2 diabetes, whereas all of our patients had insulin-dependent type 2 diabetes. Furthermore, in our study, only two of the 10 patients were able to stop insulin treatment 4 months after the bypass surgery. It can therefore not be excluded that continued hyperinsulinemia may have influenced cardiac function.
A limitation of our study is the small sample size. Because of the narrow interior of the MRI scanner and the maximum allowable weight of the MRI table, many candidates for bariatric surgery are not eligible for MR studies. However, our study was successful in showing substantial changes of bariatric surgery on ectopic fat accumulation and in particular provided new data on fat distribution in epi- and paracardial fat. Another limitation is the relatively short follow-up period, which could be an explanation for the fact that we did not find statistically significant differences in cardiovascular function parameters between the two time points.
In conclusion, this study shows that weight loss induced by bariatric surgery leads to a decrease in abdominal and pericardial fat depots, with a higher relative decrease in visceral and paracardial fat volumes. These findings contribute to the existing evidence suggesting tissue-specific changes in body fat distribution after weight loss and exercise interventions. However, the decrease in pericardial fat did not lead to improved cardiovascular function after the RYGB.