Normalization of Cardiac Function After Bariatric Surgery Is Related to Autonomic Function and Vitamin D

Purpose Subclinical cardiac dysfunction is common in patients with obesity. Bariatric surgery is associated with normalization of subclinical cardiac function in 50% of the patients with obesity. The aim of this study was to identify predictors for a lack of improvement of subclinical cardiac dysfunction 1-year post-bariatric surgery. Methods Patients who were referred for bariatric surgery were enrolled in a longitudinal study. Inclusion criteria were age 35–65 years and BMI ≥ 35 kg/m2. Patients with a suspicion of or known cardiovascular disease were excluded. Conventional and advanced echocardiography, Holter monitoring, and blood tests were performed pre- and 1-year post-bariatric surgery. Subclinical cardiac dysfunction was defined as either a reduced left ventricular ejection fraction, decreased global longitudinal strain (GLS), diastolic dysfunction, arrhythmia, or an increased BNP or hs Troponin I. Results A total of 99 patients were included of whom 59 patients had cardiac dysfunction at baseline. Seventy-two patients completed the 1-year follow-up after bariatric surgery. There was a significant reduction in weight and cardiovascular risk factors. Parameters of cardiac function, such as GLS, improved. However, in 20 patients cardiac dysfunction persisted. Multivariate analysis identified a decreased heart rate variability (which is a measure of autonomic function), and a decreased vitamin D pre-surgery as predictors for subclinical cardiac dysfunction after bariatric surgery. Conclusion Although there was an overall improvement of cardiac function 1-year post-bariatric surgery, autonomic dysfunction and a decreased vitamin D pre-bariatric surgery were predictors for a lack of improvement of subclinical cardiac dysfunction. Graphical abstract


Introduction
Obesity has reached epidemic proportions globally and the prevalence is still increasing [1]. Subclinical cardiac dysfunction is common in patients with obesity, [2] and obesity is associated with an increased risk of heart failure [3]. Heart failure is characterized by an impaired quality of life, frequent hospitalizations, and poor outcome [4]. Considering that prevention and treatment of heart failure have enormous medical and socioeconomic implications, a deeper understanding of risk factors for heart failure such as obesity is imperative.
Clinically significant weight loss is difficult to achieve with lifestyle interventions and the results are often temporary. In contrast, bariatric surgery is an effective and safe treatment option resulting in large long-term weight loss [5]. Several studies suggest that weight loss achieved by bariatric surgery has a positive impact on heart morphology, even in patients with obesity without heart failure [6]. We recently demonstrated that subclinical cardiac dysfunction normalized in half of the patients with obesity 1-year after bariatric surgery [7]. Also, bariatric surgery is associated with a 35% reduced incidence of new-onset heart failure during long term follow-up [8]. However, little is known about the pathophysiology of cardiac dysfunction in obesity patients and the factors determining the evolution of cardiac function after bariatric surgery are unknown. We have previously shown that subclinical cardiac dysfunction is related to autonomic dysfunction in obesity patients, [2] but it is unknown whether autonomic dysfunction may be related to a lack of recovery of cardiac dysfunction after bariatric surgery as well.
The CARdiac Dysfunction In Obesity -Early Signs Evaluation (CARDIOBESE) study was the first study in which (speckle tracking) echocardiography, blood tests, and Holter monitoring were combined to simultaneously investigate different aspects that may all play a role in the pathophysiology of subclinical cardiac dysfunction in obesity patients. The aim of this study was to identify predictors for persistent cardiac dysfunction 1-year post-bariatric surgery.

Study Design and Study Group
The protocol of the CARDIOBESE study has been described before [9]. In short, the CARDIOBESE study is a longitudinal study in which we prospectively enrolled 100 patients with obesity who were referred for bariatric surgery to the Franciscus Gasthuis & Vlietland (75 patients) and Maasstad Ziekenhuis (25 patients), both in Rotterdam, the Netherlands. Patients were included if they were between 35 and 65 years old and had a BMI of ≥ 35 kg/m [2]. Patients with a suspicion of or known cardiovascular disease were excluded. Bariatric surgery was performed by either a gastric sleeve, a gastric bypass or one anastomosis gastric bypass (OAGB) operation. Patients were seen pre-and 1-year post-bariatric surgery to study the intra-personal impact of obesity and bariatric surgery-related changes on cardiac function. The study protocol was approved by the ethics committee and written informed consent was obtained from all participants included in the study [9]. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
The presence or absence of subclinical cardiac dysfunction in the 100 patients with obesity of the CARDIOBESE-study has been described in detail before [2]. In short, cardiac dysfunction was defined as either a reduced LV ejection fraction, [10] a decreased global longitudinal strain (GLS) (< 17%), diastolic dysfunction [11], ventricular arrhythmia or an increased BNP (> 30 pmol/L) or hs Troponin I(≥ 34 ng/L for male and > 16 ng/L for female subjects). Of the predefined studied parameters, a decreased GLS (< 17%) was by far the most abundant, in 57 patients; one had diastolic dysfunction without an available GLS, one had a normal GLS but an increased BNP (49 pmol/L, normal value < 30 pmol/L), and one had a positive hs Troponin I. One patient with cardiac dysfunction was diagnosed with acromegaly after inclusion and was excluded from further analysis, leaving 59 patients with versus 40 without subclinical cardiac dysfunction.
To optimize speckle tracking echocardiography, apical images were obtained at a frame rate of 60 to 80 frames/s. Three consecutive cardiac cycles were acquired from all apical views. Subsequently, these cycles were transferred to a QLAB workstation (version 10.2, Philips, Best, the Netherlands) for off-line speckle tracking analysis. Peak regional longitudinal strain was measured in 17 myocardial regions and a weighted mean was used to derive GLS.

Blood Tests
Non-fasting blood samples were taken both for the study and as part of regular care. Routine laboratory measurements included; glucose, glycosylated haemoglobin (HbA1C), creatinine, estimated glomerular filtration rate (eGFR), alanine aminotransferase (ALAT), Apolipoprotein B, total cholesterol, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), triglycerides, ferritin, active vitamin B12, folic acid, vitamin B1, vitamin B6, albumin, magnesium, vitamin D, and haemoglobin were determined by standard clinical procedures as described [12]. In addition to the regular patient care path blood tests, high sensitive troponin I (hs troponin I), C reactive protein (CRP), and brain natriuretic peptide (BNP) were determined specifically for this study.

Holter Monitoring
Heart rhythm was recorded for 24 consecutive hours using a portable digital recorder (GE HEER Light, USA). The digital recorder was connected using stickers that were placed on the chest. Average heart rate, minimal heart rate, maximum heart 1 3 rate, total premature atrial contractions (PAC), total premature ventricular contractions (PVC), the standard deviation of all NN (often also referred to as RR) intervals (SDNN), and SDNN index were measured. A 24-h recording of the SDNN reveals the sympathetic nervous system contribution to heart rate variability [13]. The SDNN index estimates the variability due to the factors affecting heart rate variability (HRV) within a 5-min period. It is calculated by first dividing the 24-h record into 288 5-min-segments and then calculating the standard deviation of all NN intervals contained within each segment. [14]

Statistical Analysis
Patients who completed the follow-up were included in the analysis. The normality of the data was checked by the Shapiro-Wilk test. Continuous values with normal distributions were expressed as mean ± standard deviation, with skewed distributions as median and interquartile range and categorical values as percentages. The paired Student's t-test was used for continuous variables with normal distributions, the nonparametric Wilcoxon signed-rank test for variables with skewed distributions, and the McNemar test for categorical variables was used to compare parameters pre-and post-surgery.
The unpaired Student's t-test for continuous variables was used to compare the pre-and post-surgery values of patients with versus without cardiac dysfunction post-surgery, the non-parametric Mann-Whitney U test for continuous parameters with skewed distributions, and the χ 2 test for categorical variables. Pre-surgery parameters that significantly differed between patients with post-surgery normal cardiac function and patients with post-surgery cardiac dysfunction in the univariate analyses were added to multivariate logistic regression analysis (method: backward stepwise analysis). The discriminative ability of the resulting model was investigated by calculating the area under the receiver operating curve (AUC). Odds ratios and 95% confidence intervals were calculated. A two-tailed p-value of < 0.05 was considered statistically significant. Statistical analyses were performed with SPSS version 26.0 or higher (SPSS Inc., Chicago, USA).

Changes in Features of Obesity from Pre-to 1-Year Post-Bariatric Surgery
A total of 100 patients with obesity were included, 85 patients underwent bariatric surgery and 72 patients completed the 1-year follow-up ( Fig. 1). Fifteen patients did not undergo bariatric surgery because of various reasons, but mostly because of disapproval by the psychologist or because they withdrew from surgery for personal reasons.
In Table 1, it is shown that weight loss and decreased BMI were significant 1-year post-bariatric surgery. Systolic blood pressure and heart rate decreased significantly as well.
Also, the prevalence of comorbidities such as diabetes mellitus, hypertension, and obstructive sleep apnoea syndrome decreased significantly. Medication use was reduced postsurgery, with a significant reduction in use of ACE inhibitors/ angiotensin receptor blockers, statins, and oral anti-diabetics.
Blood tests showed a significant decrease in CRP, HbA1c, creatinine, ALAT, Apolipoprotein B, total cholesterol, LDL-C, and triglycerides post-bariatric surgery. HDL-C, folic acid, vitamin B6, and vitamin D increased significantly. The echocardiogram showed a decrease in LVM, but when corrected for the body surface area (LVM index), there was no significant decrease. Holter monitoring showed a decreased mean and minimal and maximum heart rate 1-year postsurgery, whereas the SDNN and the SDNN index increased.

Changes of Parameters of Cardiac Dysfunction from Pre-to 1-Year Post-Bariatric Surgery
There was a mild but statistically significant increase in BNP 1-year post-bariatric surgery ( Table 2). Levels of hs troponin I were comparable. Echocardiography showed a significant improvement of GLS. The prevalence of diastolic dysfunction and the LV ejection fraction did not change. Also, the frequency of extrasystoles did not change from pre-to 1-year post-bariatric surgery.

Comparison of Patients with Versus Without Normalization of Cardiac Function After Bariatric Surgery
Of the patients with complete follow-up, 40 (56%) had subclinical cardiac dysfunction pre-surgery. In 50% of these patients, cardiac function had normalized 1-year postsurgery (Table 3). In the 20 patients in whom subclinical cardiac dysfunction persisted, 17 (43%) had a decreased GLS, one patient had an elevated hs troponin I level, and two patients had diastolic dysfunction.   which were different pre-surgery, identifying a decreased SDNN and a decreased vitamin D pre-surgery as significant predictors for maintaining cardiac dysfunction after bariatric surgery (Table 4). The multivariate model including these two parameters to identify patients who maintained cardiac dysfunction post-surgery had an AUC of 0.81 (95% CI: 0.67-0.95, p = 0.001), with a sensitivity of 70% (95%CI: 66-87%) and a specificity of 80% (95%CI: 56-93%) (Fig. 2).

Discussion
The main finding of the current study is that persistence of cardiac dysfunction in patients with obesity 1 year after bariatric surgery was related to autonomic dysfunction and a decreased vitamin D pre-surgery. Although in previous studies changes of cardiac morphology and function after bariatric surgery have been investigated, [6,8] CARDIOBESE is the first study in which the focus was specifically on subclinical cardiac dysfunction. Furthermore, analysis with the combination of speckle tracking echocardiography, blood tests, and Holter monitoring was used for the first time to simultaneously investigate different aspects of cardiac dysfunction and the underlying pathophysiology. As expected, and in-line with previous findings, [6,8] many cardiovascular risk factors and parameters of cardiac function improved post-surgery. Prevalence of comorbidities decreased, lipid levels and HbA1c improved, and CRP decreased. Also, there was a mild but statistically significant increase of BNP 1-year post-surgery. BNP is known to be decreased in patients with obesity, both with and without heart failure [15]. Although the reason for this remains incompletely understood, it is most likely due to lower release in these patients, rather than increase in their clearance. [16] Improvement of LV function following bariatric surgery has been described before in small studies [17][18][19][20]. However, CARDIOBESE is the largest study in which speckle tracking echocardiography was used to investigate improvement of LV function after bariatric surgery. As we recently reported, there was an overall improvement of GLS 1-year post-surgery, resulting in normalization of subclinical cardiac dysfunction in 50% of the patients with obesity. [7] While it was already known that autonomic dysfunction as expressed by a decreased HRV may be related to either cardiac dysfunction [21] or to obesity, [22] previously reported baseline data of the patients included in the CAR-DIOBESE study [2] for the first time showed that autonomic dysfunction appears to have a prominent role in the pathophysiology of cardiac dysfunction in obesity. However, so far, it was unknown whether autonomic dysfunction may play a role in persistence of cardiac dysfunction after bariatric surgery as well. In the current study, it was shown that a decreased SDNN pre-surgery was a predictor for persistent subclinical cardiac dysfunction 1-year post-bariatric surgery. The SDNN represents the beat-to-beat variation during Holter monitoring by measuring the standard deviation of NN intervals [22]. The SDNN is a parameter of autonomic function through the sympathetic nervous system contribution to HRV [13]. A balanced autonomic function is crucial for normal cardiac function [21]. On the other hand, a depressed HRV is related to morbidity and mortality [23,24]. Oher studies already described a favourable effect of bariatric surgery on HRV [25]. Yet, by combining findings from Holter monitoring and echocardiography, our study is the first to relate the severity of autonomic dysfunction in obesity to the potential of recovery of cardiac dysfunction after bariatric surgery.
In the patients in our study, there was a significant increase in SDNN 1-year post-surgery, indicative of improvement of autonomic function, both in patients with improvement of LV function and in patients with persistent LV dysfunction. It can therefore be hypothesized that more severe autonomic dysfunction in obesity as expressed by decreased SDNN presurgery, may lead to either a permanent or delayed lack of improvement of LV function after bariatric surgery. Longer follow-up of obesity patient post-bariatric surgery may elucidate whether LV function will improve after all, in-line with improvement of autonomic function.
While as described above, a role of autonomic dysfunction was somewhat anticipated, the finding that a decreased vitamin D before bariatric surgery was also independently related to persistent subclinical cardiac dysfunction 1-year post-surgery was less expected. Nevertheless, vitamin D has been suggested to be involved in multiple pathophysiological pathways related to heart failure, such as inflammation, atherosclerosis, endothelial dysfunction, and thrombosis [26]. Furthermore, vitamin D deficiency is a predictor of reduced survival in patients with heart failure [27]. Also, vitamin D is known to be decreased in patients with obesity, [28] and in patients with known cardiovascular disease [29], suggesting that vitamin D may have a role in the increased risk of cardiac dysfunction in obesity. However, previous studies from our group failed to show significant effects of vitamin D supplementation on inflammatory changes in females with overweight, making this mechanism less likely [30]. Although the underlying mechanism remains to be elucidated, by combining findings from blood tests and echocardiography in our study, it was shown for the first time that a relative decreased vitamin D level pre-bariatric surgery is related to a lack of improvement of cardiac function after bariatric surgery.

Limitations
A relatively large number (32%) of the patients with cardiac dysfunction did not complete the follow-up: 15% because they did not undergo bariatric surgery, and 17% dropped out because of various other reasons. Meanwhile, 20% of the patients with a normal cardiac function was lost to   follow-up. The reason for this difference is unknown, but probably it was just coincidence. Furthermore, follow-up after bariatric surgery was 1 year and it may be hypothesized that a longer follow-up would have shown improvement of cardiac function in a larger proportion of patients.

Conclusions
Autonomic dysfunction at baseline was related to a lack of normalization of cardiac function in patients with obesity 1 year after bariatric surgery. This result is in-line with previous findings of our group, [2] confirming an important role of autonomic dysfunction in the pathophysiology of cardiac dysfunction in obesity. Decreased vitamin D before bariatric surgery was also independently related to persistent subclinical cardiac dysfunction 1-year post-surgery. Since this finding was less expected, we consider this less affirmative and more hypothesis-generating. Nevertheless, signs of either autonomic dysfunction or a decreased vitamin D pre-bariatric surgery may be indicative of a need for cardiologic follow-up after bariatric surgery.
Funding This work was supported by a grant from Stichting BeterKeten.

Conflict of Interest
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.