Highlights
- •
Significant reductions in oral diuretic use were observed, with an average reduction of 65% at 24 months after metabolic surgery.
- •
Decrease in body mass index and total body weight loss observed at 24 months are greater than reported in the general population.
- •
Improvements in hemoglobin A1c levels were sustained ≥24 months postoperatively.
- •
Trends toward less emergency department utilization for cardiac conditions and intravenous diuresis were observed.
- •
Metabolic surgery indicates promising metabolic and cardiac benefits in patients with heart failure.
The beneficial impacts of metabolic surgery (MS) on patients with heart failure (HF) are incompletely characterized. We aimed to describe the cardiac and metabolic effects of MS in patients with HF and hypothesized that patients with HF would experience both improved metabolic and HF profiles using glycemic control and diuretic dependency as surrogate markers. In this single-center, university-affiliated academic study in the United States, a review of 2,342 hospital records of patients who underwent MS (2017 to 2023) identified 63 patients with a medical history of HF. Preoperative characteristics, 30-day outcomes, and up to 2-year biometric and metabolic outcomes, medication usage, and emergency department utilization were collected. At 24 months, mean body mass index change was −16 kg/m 2 (p <0.001) that corresponded to a mean percentage total body weight loss of 29% (p <0.001). Weight loss was accompanied by significant reductions in hemoglobin A1c (p <0.001) and a 65% decrease in diuretic use at 24 months after surgery (p <0.001). Similarly, emergency visits for cardiac conditions (p = 0.06) and intravenous diuresis (p = 0.07) trended favorably at 1 year after surgery compared with 1 year before surgery but were not statistically significant. In conclusion, in patients with HF who were carefully selected, MS appears to provide significant reduction in oral diuretic dependency, and metabolic improvements with trends toward lower rates of emergency department utilization.
Obesity is a widely recognized co-morbid condition of heart failure (HF), with strong evidence linking obesity as a driver of HF development. A dose-dependent relation indicates that each incremental increase in body mass index (BMI) significantly elevates the risk, especially in HF with preserved ejection fraction (HFpEF) compared with HF with reduced ejection fraction (HFrEF). , Weight reduction with metabolic surgery (MS), also known as metabolic and bariatric surgery, has been consistently shown to reduce new-onset HF. Specifically, the Atherosclerosis Risk in Communities (ARIC) study revealed a 46% greater risk of HF incidence for every 5-kg/m 2 increase in BMI. Conversely, Aminian et al showed a 62% smaller risk of HF development at any given time for patients who underwent MS than for a matched group of nonsurgical controls. However, the effect of MS on preexisting HF is less well understood. HF in the setting of co-morbid obesity is strongly associated with negative effects on both symptom burden and exercise intolerance. , Weight reduction in the setting of HF has been proposed to increase exercise tolerance and improve quality of life, with retrospective cohorts suggesting significant improvements in mortality. Despite this, the measurable effects of MS on preexisting HF, namely, diuretic use and emergency room utilization, remain unclear. Using diuretic dependency and emergency visit utilization as surrogates for clinical HF improvement, this study aims to describe the cardiac and metabolic outcomes of MS in patients with HF, hypothesizing improved metabolic and HF profiles and acceptable 30-day complication rates.
Methods
A retrospective analysis was conducted using electronic health record data from a single academic health center in addition to variables previously abstracted for submission to the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program. After approval by the local institutional review board, MS cases occurring between January 2017 and December 2023 were extracted, producing a total of 2,342 cases. Using SlicerDicer (vMay2023, Epic Systems, Verona, Wisconsin) for extraction, cases were queried for problem list diagnoses of HF corresponding to International Classification of Diseases code I50. After excluding duplicate results, 63 records were identified.
Automated clinical data pulls were performed using Research Electronic Data Capture (v14.1.2 Vanderbilt University, Nashville, Tennessee), a secure, web-based software platform designed to support data capture, to serially collect preoperative and 1-, 6-, 12-, 18-, and 24-month postoperative data for weight (kg), hemoglobin A1c (HbA1c), basic metabolic panel (BMP), and B-type natriuretic peptide as recorded during the course of usual postoperative care. Patient charts were also manually reviewed at preoperative and 1-, 6-, 12-, 18-, and 24-month postoperative time points to record medication changes, emergency department (ED) visits, and cardiac testing to attain ejection fraction. Specifically, prescription oral diuretic dosage was recorded at each time point and converted into daily furosemide equivalents (FEs) for analysis ( Figure 1 ). All in-system instances of emergency visits were assessed for use of intravenous diuresis or cardiac condition/diagnosis for the 12 months before MS intervention and at 12-month intervals after intervention.
Thirty-day Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program variables—including patient characteristics, surgical details, co-morbid conditions, and postoperative outcomes—were matched to local records. Variables included age, gender, race, procedure type, procedure approach, smoking history, functional status, diabetes mellitus, hyperlipidemia, hypertension, history of myocardial infarction (MI), previous cardiac catheterization, preoperative obstructive sleep apnea, history of gastroesophageal reflux disease, history of severe chronic obstructive pulmonary disease, history of venous thromboembolism, history of pulmonary embolism (PE), preoperative venous stasis, preoperative inferior vena cava (IVC) filters, currently requiring dialysis preoperatively, preoperative renal insufficiency, average hospital stay, unplanned reintubation, discharge destination, 30-day readmissions, 30-day superficial incisional surgical site infections (SSIs), 30-day pneumonia, 30-day PE, 30-day organ space SSI, 30-day cardiac arrest requiring cardiopulmonary resuscitation, 30-day deep incisional SSI, 30-day wound disruption, 30-day venous thrombosis requiring therapy, 30-day sepsis, 30-day septic shock, 30-day acute renal failure, and 30-day stroke.
Data were cleaned, processed, and analyzed in R (v4.3.1, R Foundation for Statistical Computing, Vienna, Austria). In addition, hmisc, tidyverse, gridExtra, and geepack packages were used for data manipulation, statistical analysis, and visualization. We used Locally Estimated Scatterplot Smoothing for the visualization of data trends presented in Figures 2 and 3 . For the analysis of variations in longitudinal data while considering the correlation of repeated measurements within subjects, we applied regression models incorporating discrete time points through Generalized Estimating Equations with an exchangeable correlation structure. To assess statistical differences in anthropometric measures, HbA1c levels, and FEs, we used analysis of variance techniques. Moreover, Wilcoxon signed-rank tests facilitated the comparison of variables in 12-month preoperative and postoperative periods.
Results
The study cohort comprised 63 patients ( Table 1 ). Age at the time of operation ranged from 28 to 69 years, with mean and median ages of 53 and 53 years, respectively. Most of the patients, 43 (68%), were female, and most (n = 53, 88%) had HFpEF compared with 5 (8%) with HFrEF. The racial composition was predominantly White (n = 35, 56%), with Black or African-American making up another 43% (n = 27).
| Baseline Patient Characteristics: | 30-day Complications: | ||
|---|---|---|---|
| HF, n (%) | 63 (100%) | ||
| HFpEF | 53 (88%) | ||
| HFrEF | 5 (8.3%) | Readmission, n (%) | 9 (14%) |
| Unknown | 3 (3.3%) | Reoperation, n (%) | 4 (6.4%) |
| Age, median (IQR) | 53 (40-67) | Nonoperative Intervention, n (%) | 4 (6.4%) |
| BMI, median (IQR) | 52 (37-68) | Unplanned ICU Admit, n (%) | 5 (7.9%) |
| Race, n (%) | |||
| White | 35 (56%) | ||
| Black | 27 (43%) | Surgical Site Infection, n (%) | |
| Other/Unknown | 1 (1.5%) | Superficial Incisional | 2 (3.2%) |
| Gender, n (%) | Deep Incisional | 0 (0%) | |
| Female | 43 (68%) | Sepsis, n (%) | 0 (0%) |
| Male | 20 (32%) | ||
| Diabetes, n (%) | Pneumonia, n (%) | 1 (1.6%) | |
| No | 29 (46%) | Urinary Tract Infection, n (%) | 1 (1.6%) |
| Yes, non-insulin | 21 (33%) | ||
| Yes, insulin | 13 (21%) | ||
| COPD, n (%) | 1 (1.6%) | Cardiac Arrest, n (%) | 2 (3.2%) |
| MI History, n (%) | 5 (7.9%) | Acute Renal Failure, n (%) | 0 (0%) |
| Hypertension, n (%) | 57 (91%) | ||
| Hyperlipidemia, n (%) | 45 (71%) | ||
| Renal Insufficiency, n (%) | 6,143 (0.6%) | ||
| Cardiac Intervention, n (%) | 2 (3.2%) | ||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
