Effects of bariatric surgery on cardiovascular‐related acute care use in patients with hypertrophic cardiomyopathy

Abstract

Aims

Prior studies have suggested causal relationships between obesity and acute cardiovascular events. It has been also known that the risk of acute cardiovascular events is reduced by bariatric surgery. However, little is known about whether bariatric surgery lowers the risk of acute cardiovascular events in patients with obesity and hypertrophic cardiomyopathy (HCM). In this context, we aimed to investigate whether bariatric surgery is associated with a reduced risk of cardiovascular‐related acute care use in patients with HCM.

Methods and results

In this population‐based study, the bariatric surgery group consisted of patients with HCM who underwent bariatric surgery from January 2004 to December 2014. The control group included those who have obesity and HCM and received non‐bariatric elective intra‐abdominal surgery during the same period. The outcome was cardiovascular‐related acute care use (i.e. emergency department (ED) visits or unplanned hospitalizations for cardiovascular disease) during a 1‐year post‐surgery period. We used the SPARCS database, a population‐based ED and inpatient database in New York State. We constructed logistic regression models with generalized estimating equations to compare the risk of the outcome events during sequential 6‐month post‐surgery periods. We adjusted for age, sex, number of ED visits and hospitalizations for cardiovascular disease within 2 years before the index surgery, and the Elixhauser co‐morbidity measures. We also performed propensity score (PS)‐matching and inverse probability treatment weighting analyses using these variables. The analytic cohort consisted of 207 adults with obesity and HCM, including 147 patients who underwent bariatric surgery and 60 in the control group. The risk was not significantly different in the 1–6 months post‐surgery period. By contrast, in the 7–12 months post‐surgery period, the risk of cardiovascular‐related acute care use was significantly lower in the bariatric surgery group (adjusted odds ratio 0.23; 95% CI 0.068–0.71; P = 0.01) compared with the control group. In the PS‐matched cohort, there were no significant differences in the baseline characteristics. The PS‐matched analysis demonstrated lower risk of the outcome event in the bariatric surgery group in the 7–12 months post‐surgery period. The inverse probability treatment weighting analysis replicated the findings.

Conclusions

Bariatric surgery was associated with a lower risk of cardiovascular‐related acute care use in the 7–12 months post‐surgery period in this population‐based study.

Introduction

Hypertrophic cardiomyopathy (HCM) is the most common genetic heart disease in the United States. ¹ , ² A complicated interaction between genotype and environmental factors determines phenotype and clinical course of the disease. ³ , ⁴ Obesity is one of the most important modifiable risk factors for a variety of cardiovascular diseases such as dyslipidemia, type 2 diabetes mellitus, hypertension, and heart failure. ⁵ HCM is not an exception—a retrospective study demonstrated that body mass index (BMI) was associated with the magnitude of hypertrophy in HCM. ⁶ Another cohort study revealed that obesity was associated with HCM‐related morbidities (e.g. ventricular arrhythmia, heart failure, atrial fibrillation, and stroke). ⁶ , ⁷ As such, the current guideline recommends weight loss in patients with obesity and HCM in an attempt to avoid unfavourable cardiovascular events. ⁸

Bariatric surgery is the most effective method to achieve substantial weight loss. ⁹ Prior studies in the non‐HCM population have reported that significant weight loss is associated with a decreased risk of acute cardiovascular events such as myocardial infarction, stroke, heart failure exacerbation, angina, and hypertensive crisis. ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ However, there is a lack of evidence on the association between weight loss and the risk of acute cardiovascular events in patients with obesity and HCM. In this context, we aimed to investigate whether bariatric surgery, used as an instrument to achieve substantial and sustained weight loss, is associated with a reduced risk of cardiovascular events requiring acute care use in patients with obesity and HCM.

Methods

Study design and setting

We analysed a population‐based data to evaluate the effects of bariatric surgery on the risk of cardiovascular‐related acute care use, defined as emergency department (ED) visit or unplanned hospitalization for cardiovascular disease (CVD), in patients with obesity and HCM. We used a population‐based database—the Statewide Planning and Research Cooperative System (SPARCS)—from 2004 to 2014. ¹⁷ SPARCS is a comprehensive all‐payer data reporting system. ¹⁷ It includes patient‐level data on patient characteristics, diagnoses and treatments, services, and charges for every inpatient stay and outpatient visit (i.e. ambulatory surgery, ED, and outpatient services) in New York State. ¹⁷ SPARCS was established in 1978 and each discharge has been associated with a unique patient identifier, thereby providing one of the longest longitudinal information on patient outcomes in New York State. ¹⁸ Details of the database have been reported previously. ¹⁹ , ²⁰ , ²¹ The institutional review board of Columbia University Irving Medical Center approved this study.

Study population

We took the following steps to include all adult patients with obesity and HCM who underwent bariatric surgery (i.e. the bariatric surgery group) and non‐bariatric elective intra‐abdominal surgery (i.e. the control group). We chose this control group because patients who underwent non‐bariatric elective intra‐abdominal surgery would experience similar post‐surgical changes as bariatric surgery without intervening on weight (Table  S1 ).

First, we identified adults (aged ≥18 years) with obesity and HCM, which was defined by having at least one encounter (including inpatient stay and outpatient visit) with a diagnostic code for HCM and at least one encounter with a diagnostic code for obesity in any of the diagnostic code fields during 1 January 2004 to 31 December 2014. HCM was defined by International Classification of Diseases, Ninth Revision, Clinical Modification (ICD‐9‐CM) diagnosis code 425.1x or International Classification of Diseases, Tenth Revision, Clinical Modification (ICD‐10‐CM) diagnosis code I42.1 and I42.2. ²² We further identified patients with obstructive HCM by using ICD‐9 diagnostic code 425.11 and ICD‐10 diagnostic code I42.1. Obesity was defined by the ICD‐9‐CM diagnostic codes 278.0 to 278.2, V77.8, V85.3x, and V85.4, and the corresponding ICD‐10‐CM diagnostic codes E66, E66.0, E66.1, E66.2, E66.8, E66.9, and Z68.4. ¹⁰ , ¹⁴ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸

Second, as the bariatric surgery group, we specified patients with obesity and HCM who had bariatric surgery between January 1, 2004 and December 31, 2014. We used ICD‐9‐CM and ICD‐10 procedure codes to define bariatric surgery as shown in Table S2 . ¹⁶ , ²⁹ , ³⁰ We also determined subtypes of bariatric surgery—that is, gastric bypass surgery, gastric sleeve surgery, and other bariatric surgery—by using ICD‐9 and ICD‐10 codes.

Third, as the control group, we further identified patients with obesity and HCM who never underwent bariatric surgery but underwent non‐bariatric elective intra‐abdominal surgery during 1 January 2004 to 31 December 2014. We defined non‐bariatric elective intra‐abdominal surgery by the ICD‐9‐CM procedure codes and the corresponding ICD‐10 procedure codes as listed in Table S1 . ³¹ , ³²

Last, we excluded patients who underwent multiple elective surgeries from both groups. ¹⁰ , ²³ , ³⁰ , ³³ We also excluded patients with diagnostic codes for gastrointestinal cancer (ICD‐9: 150.0 to 159.9; ICD‐10: C15‐C26) from both groups. ¹⁰ , ²³ , ³⁰ , ³³ We defined the starting date of follow‐up as the day of the surgery in both groups.

Measurements and outcomes

We retrieved data from the SPARCS database on the patient demographics (age, sex, and race/ethnicity), source of payment (Medicare, Medicaid, private insurance, self‐pay, or other), ICD‐9‐CM diagnosis and procedure codes, ICD‐10‐CM diagnosis and procedure codes, and co‐morbidities defined by Elixhauser co‐morbidity measures. ¹⁰ , ¹⁴ , ²³ , ³⁰ , ³⁴ We used the baseline characteristics information recorded during the index hospitalization for either bariatric surgery or elective intra‐abdominal surgery.

The primary outcome measure was cardiovascular‐related acute care use, defined as ED visit or unplanned hospitalization with the primary diagnosis of CVD—that is, the Clinical Classifications Software codes 96–121 in the primary diagnosis field. ³⁰

Statistical analysis

For comparisons of the baseline characteristics between patients in the bariatric surgery group and the control group, Student's t‐test or the χ ² test was used as appropriate. The number of patients and risk of the primary outcome event were calculated for 1–6 months and 7–12 months after surgery. The risk of the corresponding periods was compared between patients in the bariatric surgery group and those in the control group by fitting logistic regression models with generalized estimating equations. We compared the risk of cardiovascular‐related acute care use between the two groups using univariable analysis as well as multivariable models adjusting for age, sex, race/ethnicity, source of payment, hospital site, the number of prior ED visit or hospitalization for CVD within 2 years before the surgery, and Elixhauser co‐morbidity measures.

In addition, we conducted propensity score (PS)‐matched analysis. PS‐matching methods provide complementary advantages over traditional multivariable analysis to control for confounding by indication. ³⁵ , ³⁶ We used PS to match patients based on a set of covariates so that each group has similar distribution of potential confounders. While PS‐matching may lead to a smaller sample size, it provides a clinically relevant estimate of the effects because patients in the matched sample are potential candidates for either bariatric surgery or non‐bariatric elective intra‐abdominal surgery if the clinical need arises. ²² We computed PS with the use of a logistic regression model to estimate the propensity that patients would undergo bariatric surgery. In the propensity model, we included age, sex, race/ethnicity, source of payment, the number of prior ED visit or hospitalization for CVD within 2 years before the index surgery, and Elixhauser co‐morbidity measures. We matched patients in the bariatric surgery group to those in the control group according to PS at a 2:1 ratio. We performed the matching without replacement, by using callipers (width Z = 0.1) of the standard deviation of the logit of the PS.

Finally, we applied an inverse probability of treatment weighting (IPTW) analysis. Akin to the PS‐matched analysis, we computed IPTW with the use of a logistic regression model to estimate the propensity that patients would undergo bariatric surgery. In the IPTW model, we included the same parameters as the PS‐matched analysis. We applied stabilized weights, based on the inverse of the PS, to generate a weighted cohort in which covariate distributions were independent of treatment assignment. ³⁷

In the univariable, multivariable, PS‐matched, and IPTW models, odds ratio (OR) of cardiovascular‐related acute care use was estimated. The ORs were computed by using logistic regression models with the control group as the reference. All analyses were performed at a 2‐sided significance level of 0.05, and all confidence intervals (CIs) were reported as two‐sided values with a confidence level of 95%. Statistical analyses were performed using STATA 14.1 (StataCorp; College Station, TX).

Results

The analytic cohort consisted of 207 adults with obesity and HCM, including 147 patients in the bariatric surgery group (18 patients with gastric bypass surgery, 72 with gastric sleeve surgery, and 57 with other bariatric surgery) and 60 patients in the control group. The baseline characteristics are described in Table 1 . At baseline, patients who underwent bariatric surgery were significantly older (P = 0.01) and had a significantly higher prevalence of female (P = 0.02). In this cohort, 100 patients had obstructive HCM and 107 patients had non‐obstructive HCM. There was no death immediately following bariatric surgery or non‐bariatric elective intra‐abdominal surgery. At 1‐year follow‐up, six (10%) patients in the bariatric surgery group and 14 (9.5%) patients in the control group died (P > 0.99).

Table 1.

Baseline characteristics of adults with obesity and hypertrophic cardiomyopathy who underwent bariatric surgery and those in the control group who had obesity and non‐bariatric elective intra‐abdominal surgery

Characteristics	Bariatric surgery	Control	P value
	(n = 147)	(n = 60)
Age ± standard deviation (year)	63 ± 13	57 ± 14	0.01
Female sex	74 (49%)	41 (68%)	0.02
Race/ethnicity			0.16
Non‐Hispanic white	75 (52%)	32 (53%)
Non‐Hispanic black	37 (26%)	15 (25%)
Hispanic	11 (8%)	0 (0%)
Asian	0 (0%)	0 (0%)
Other	21 (15%)	13 (22%)
Primary insurance			0.16
Medicare	78 (54%)	28 (47%)
Medicaid	14 (10%)	8 (13%)
Private	18 (13%)	13 (22%)
Self‐pay	0 (0%)	1 (2%)
Other	34 (24%)	10 (17%)
Selected co‐morbidities
Heart failure	38 (26%)	16 (27%)	>0.99
Arrhythmia	38 (26%)	13 (21%)	0.65
Valvular disease	13 (9%)	4 (7%)	0.81
Pulmonary circulation disorder	7 (5%)	4 (7%)	0.83
Peripheral vascular disorder	10 (7%)	1 (2%)	0.25
Hypertension	78 (53%)	37 (62%)	0.33
Chronic pulmonary disease	37 (25%)	14 (23%)	0.92
Diabetes mellitus	52 (35%)	16 (27%)	0.30
Fluid and electrolyte disorders	17 (12%)	4 (7%)	0.42

The number and the risk of cardiovascular‐related acute care use according to the type of surgery are displayed in Figure 1 . The univariable analysis showed that the risk did not significantly change in the 1–6 months post‐surgery period (OR 1.11; 95% CI 0.49–2.66; P = 0.81). By contrast, in the 7–12 months post‐surgery period, the risk of acute cardiovascular events was significantly lower in the bariatric surgery group (OR 0.41; 95% CI 0.18–0.94; P = 0.03) compared with the control group. The risk was not significantly different between the subgroups, that is, obstructive versus non‐obstructive HCM (Table  S3 ) and gastric bypass versus gastric sleeve versus other types of bariatric surgery (Table  S4 ).

Figure 1.

Cardiovascular‐related acute care use in 6‐month intervals among patients with hypertrophic cardiomyopathy who underwent bariatric surgery and those in the control group who had obesity and non‐bariatric elective intra‐abdominal surgery. Shown are the number of patients with an ED visit or unplanned hospitalization for CVD, risk of the outcome event, and OR in 6‐month intervals, using unadjusted model, multivariable model (adjusting for age, sex, number of ED visits and hospitalizations for cardiovascular disease within 2 years before the index surgery, and the Elixhauser co‐morbidity measures), PS‐matching, and IPTW. ORs were estimated for each 6‐month period with control group as the reference. Co‐morbidities with prevalence <0.5% in both groups were excluded from the multivariable model. CI, confidence interval; CVD, cardiovascular disease; ED, emergency department; IPTW, inverse probability weighting; OR, odds ratio; PS, propensity score.

The multivariable analysis adjusting for the potential confounders (Figure  1 ) showed results similar to those in the univariable analysis. There were no significant differences in the risk of cardiovascular‐related acute care use between the two groups during the first 6 months. By contrast, patients who received bariatric surgery had a significantly lower risk of cardiovascular‐related acute care use in the 7 to 12 months post‐surgery period (adjusted OR 0.23; 95% CI 0.068–0.71; P = 0.01) compared with the control group.

In the PS‐matched cohort, the baseline characteristics of the two patient groups were all balanced as indicated in Table 2 (P > 0.50 for all comparisons). Consistent with the univariable and multivariable analyses, patients who underwent bariatric surgery had a significantly lower risk of cardiovascular‐related acute care use during 7 to 12 months post‐surgery period (OR 0.26; 95% CI 0.083–0.73; P = 0.01), while no difference was observed during 1 to 6 months post‐surgery period. The IPTW analysis also replicated the findings (OR 0.33; 95% CI 0.16–0.71; P = 0.004 during the 7–12 months post‐surgery period) (Figure  1 ).

Table 2.

Baseline characteristics of adults with obesity and hypertrophic cardiomyopathy who underwent bariatric surgery and 2:1 propensity score‐matched patients in the control group who had obesity and non‐bariatric elective intra‐abdominal surgery

Characteristics	Bariatric surgery	Control	P value
	(n = 82)	(n = 47)
Age ± standard deviation (year)	61 ± 14	61 ± 12	>0.99
Female sex	47 (57%)	28 (60%)	0.95
Selected co‐morbidities
Heart failure	23 (28%)	11 (23%)	0.71
Arrhythmia	20 (24%)	10 (21%)	0.85
Valvular disease	8 (10%)	3 (6%)	>0.99
Pulmonary circulation disorder	3 (4%)	1 (2%)	>0.99
Peripheral vascular disorder	3 (4%)	1 (2%)	>0.99
Hypertension	48 (59%)	29 (62%)	0.87
Chronic pulmonary disease	18 (22%)	10 (21%)	>0.99
Diabetes mellitus	26 (32%)	15 (32%)	>0.99
Fluid and electrolyte disorders	9 (11%)	3 (6%)	0.58

Discussion

Principal findings

In this population‐based study using data of 207 adults with obesity and HCM, patients who underwent bariatric surgery had a significantly lower risk of cardiovascular‐related acute care use in the 7–12 months after surgery compared with the control group. The difference was maintained in both univariable and multivariable models and was consistently observed with PS‐matching and IPTW analyses. The present study adds to the body of knowledge by demonstrating, for the first time, the effectiveness of bariatric surgery on cardiovascular‐related acute care use in patients with obesity and HCM.

Results in context

Bariatric surgery is the most effective method to achieve substantial weight loss. ⁹ Effects of significant weight loss achieved by bariatric surgery on CVD risk factors have been mainly investigated in the non‐HCM populations. According to the prior studies, the percentage of complete resolution was 77% in diabetes mellitus and 62% in hypertension, and that of substantial improvement in hyperlipidaemia was 70%. ³⁸ , ³⁹ These significant CVD risk modifications by bariatric surgery appear to translate into the risk reduction in acute cardiovascular events. In a large prospective cohort study, bariatric surgery was shown to reduce the composite of mortality, myocardial infarction, and stroke, compared with usual care for obesity. ⁴⁰ Our recent studies have also demonstrated that bariatric surgery was associated with a decline in the risk of heart failure exacerbation, hospitalizations for chronic stable angina pectoris, and hypertensive crisis in non‐HCM populations with obesity. ¹⁰ , ¹⁴ , ¹⁶ Cardiovascular‐related acute care use was reduced with bariatric surgery in patients with obesity and HCM in the present study, which is in line with the previous studies in non‐HCM populations.

In the recent years, interest on clinical significance of obesity‐associated acute cardiovascular events in HCM has been growing. In comparison to patients without HCM, patients with HCM tend to be more obese. ⁷ Moreover, among patients with HCM, patients with obesity are more likely to develop acute cardiovascular events. ² , ⁴ For example, a retrospective study of 275 adult patients with HCM examined the relationship between BMI and left ventricular mass and demonstrated that BMI was independently associated with the magnitude of left ventricular hypertrophy. ⁶ Other HCM cohort studies revealed that obesity was independently associated with HCM‐related morbidities such as ventricular arrhythmia, heart failure, atrial fibrillation, stroke, and all‐cause mortality. ⁶ , ⁷ Despite the apparent importance, there have been no studies that examined the effects of bariatric surgery, the most effective method of weight reduction, on cardiovascular events in patients with HCM. In this context, the current study provides additional value by exhibiting that bariatric surgery does not increase the risk of immediate postoperative cardiovascular events and has favourable effects to prevent cardiovascular‐related acute care use in the longer term in patients with HCM.

Cardiovascular‐related acute care use (i.e. a composite of ED visit and unplanned hospitalization for CVD) is an established outcome measure and has been used in multiple prior studies. ¹⁰ , ¹² , ¹⁴ , ¹⁶ Both ED visit and unplanned hospitalization are clinically important outcomes because these outcomes are associated with worse quality of life and prognosis. ¹⁰ , ¹² , ¹⁴ , ¹⁶ By reporting ED visits and unplanned hospitalization, the present study provides additional insights from socioeconomical perspective as well, which would be informative from the standpoint of patients and families.

HCM can cause sudden cardiac death, heart failure death, and death due to stroke. ⁴¹ Approximately 10% of patients died within 1 year after surgery in the present study. This is a higher mortality rate than what has been reported. For example, in a large retrospective study of patients with HCM, the mortality rate was 12.4% at a median follow‐up of 6.2 years. ⁷ As obesity is known to be associated with worse clinical outcomes in patients with HCM, the higher mortality rate in the present study may be attributable to the inclusion criteria that all patients in our study had obesity and underwent intra‐abdominal surgery.

Potential mechanisms

Our findings suggest that weight reduction achieved by bariatric surgery offers cardiovascular protection in patients with obesity and HCM. In addition to the aforementioned favourable effects of bariatric surgery observed in non‐HCM populations, a potential mechanism unique to HCM is a direct effect of weight loss on cardiac remodelling and changes in haemodynamics. For instance, a recent pilot trial examined the effect of the combination of Mediterranean diet and anaerobic exercise on weight loss and clinical status at 12 and 24 months in 20 patients with obesity and HCM. ⁴² The study revealed that these interventions were associated with a significant reverse remodelling of the left atrium along with an improvement of wedge and systolic pulmonary pressures as well as exercise capacity. These effects would lead to a better control of atrial fibrillation and heart failure, thus providing potential explanations for our observations in the present study.

Another possible mechanism is favourable effects of bariatric surgery on myocardial perfusion. Subendocardial ischaemia is an especially important cause of chronic angina and acute coronary syndrome without obstructive coronary artery in patients with HCM. Small vessel disease has been suggested to contribute to myocardial ischaemia and coronary microvascular dysfunction in the HCM population. ⁴³ Mechanisms that are unique to HCM include (i) increased left ventricular wall thickness and (ii) increased left ventricular intracavitary pressures due to outflow or mid‐cavitary obstruction. ⁴⁴ Indeed, acute coronary syndrome without obstructive coronary artery is not an uncommon initial presentation (prevalence 9–14%) in patients with HCM. ⁴⁵ Macrovascular disease is also common in HCM with a prevalence of up to 53%. ⁴⁶ , ⁴⁷ Bariatric surgery can theoretically improve control of chronic stable angina and prevent acute coronary syndrome in HCM via addressing risk factors of both microvascular and macrovascular disease (e.g. better lipid, hypertension, and diabetes mellitus control). ¹⁴ , ⁴⁸ This is another potential mechanism through which bariatric surgery reduces cardiovascular‐related acute care use in patients with obesity and HCM.

Advantages of the study design and methods

The PS‐matching and IPTW analyses augment the internal validity as they reduce inter‐group differences at baseline and enable a more precise examination of effects of bariatric surgery. ⁴⁹ With regard to the external validity, it has been reported that subjects participating in randomized controlled trials may be highly selected or behave differently compared with the general populations in the real‐world setting. ⁵⁰ For instance, most previously published randomized controlled trials on bariatric surgery enrolled <10% of the patients screened for eligibility. ⁵¹ By contrast, the external validity of the present study is strengthened because the SPARCS database captured all ED visits and hospitalizations in New York State, thereby allowing for a collection of large general population‐based data of patient care in the natural setting.

Potential limitations

The present study has potential limitations. First, the retrospective nature of this study may have led to potential misclassification bias in the diagnosis, treatment allocation, and outcome. However, the quality of the SPARCS database has been extensively tested in previous studies. ¹⁹ , ²⁰ , ⁵² Moreover, to ensure the accuracy of data and to avoid misclassification, the SPARCS database is routinely inspected for quality by the New York State Department of Health, in collaboration with the Vital Statistics Birth Registry, the Vital Statistics Death Registry, and trend analysis conducted by the Bureau of Biometrics. ⁵³ Second, the SPARCS database does not contain information on weight, preoperative cardiac haemodynamics, medications, severity of HCM, and other types of interventions for weight reduction such as medication and diet. Thus, there may be unmeasured confounders that are not included in the model. Third, our follow‐up period was 12 months because most significant weight changes following bariatric surgery occur within 12 months and weight stabilizes in a majority of patients thereafter. ⁵⁴ However, longer follow‐up is ultimately warranted to establish long‐term effects of bariatric surgery on cardiovascular‐related acute care use in patients with HCM. Last, because SPARCS database does not provide BMI at baseline, we used ICD‐9 and ICD‐10 codes to define obesity.

Conclusions

This population‐based cohort study serves as the first to investigate the effects of bariatric surgery on cardiovascular‐related acute care use in patients with obesity and HCM, displaying a significantly lower risk in the 7–12 months after the surgery compared with patients with obesity and HCM who did not undergo bariatric surgery. For clinicians and patients, these findings suggest that bariatric surgery not only carries weight reduction benefit but also prevents cardiovascular‐related acute care use among patients with obesity and HCM. For researchers, the present study would prompt further investigation on the mechanisms through which bariatric surgery reduces the risk of cardiovascular‐related acute care use in patients with obesity and HCM.

Funding

YJS was supported by research grants from the National Institute of Health (Bethesda, MD; R01‐HL157216), American Heart Association (Dallas, TX) National Clinical and Population Research Awards and Career Development Award, The Feldstein Medical Foundation (Clifton, NJ) Medical Research Grant, Korea Institute of Oriental Medicine (Daejeon, Republic of Korea), and Columbia University (New York, NY) Irving Medical Center Irving Institute for Clinical & Translational Research Precision Medicine Pilot Award as well as Columbia University Irving Medical Center Marjorie and Lewis Katz Cardiovascular Research Award.

Conflict of interest

None declared.

Supporting information

Table S1. ICD‐9‐CM and ICD‐10 procedure codes to define elective intra‐abdominal surgery.

Table S2. ICD‐9‐CM and ICD‐10 procedure codes to define bariatric surgery.

Table S3. Risk of the outcome event by the type of hypertrophic cardiomyopathy (obstructive vs. non‐obstructive).

Table S4. Risk of the outcome event by the type of bariatric surgery.

Miyashita, S. , Akita, K. , Zhao, Y. , Hasegawa, K. , Maurer, M. S. , Weiner, S. D. , Reilly, M. P. , Takayama, H. , and Shimada, Y. J. (2023) Effects of bariatric surgery on cardiovascular‐related acute care use in patients with hypertrophic cardiomyopathy. ESC Heart Failure, 10: 2438–2446. 10.1002/ehf2.14414.