Factors and Outcomes Associated With Cardiac Complications Among Patients Undergoing Elective Major Abdominal Surgery

Abstract

Background

Among patients undergoing major abdominal surgery (MAS), ∼3% develop cardiac complications (CC) and face poorer prognosis. This study aimed to characterize outcomes and identify factors associated with CC following MAS.

Methods

All elective adult (>17) hospitalizations for MAS (colectomy, esophagectomy, gastrectomy, hepatectomy, nephrectomy, pancreatectomy, splenectomy) were identified in the 2016-2022 National Inpatient Sample, using survey weights to generate nationally representative estimates. The primary outcome of interest was the development of CC (acute myocardial infarction (AMI) and cardiac arrest). We also evaluated patient and institutional factors associated with failure-to-rescue (FTR) following CC. Risk-adjusted analyses with multivariable regressions were used to characterize factors associated with the development of CC. Subgroup analyses were conducted for isolated AMI and cardiac arrest.

Results

Of an estimated 904 270 patients, 1.8% developed CC. Compared to others, CC were older (71 vs 62 years), less commonly female (36.5 vs 52.1%, P < 0.001), and had a higher burden of comorbidities (Elixhauser: 5 vs 3). Following risk-adjustment, older age (adjusted odds ratio (AOR) 1.02, 95% confidence interval (CI) 1.02-1.03) and higher burden of comorbidity (AOR 1.62, 95% CI: 1.59-1.66) were independently associated with CC (P < 0.05). Greater annual institution PCI and MAS caseloads were independently associated with a reduced risk of CC (P < 0.001). Furthermore, CC was associated with greater mortality (AOR 10.53, 95% CI: 8.90-12.46), respiratory complications (AOR 3.53, 95% CI: 3.18-3.91), and higher costs (β +$8,500, 95% CI: 7400-9700). On subgroup analysis, cardiac arrest revealed markedly higher mortality risk (AOR 6.86) than AMI alone (AOR 1.28).

Discussion

In summary, CC was associated with inferior outcomes and higher resource utilization. Furthermore, we found patient and hospital factors to be independently linked with CC risk. These findings highlight an association between institutional MAS and PCI volume and reduced CC risk, warranting further investigation into the role of center-level factors in perioperative cardiac outcomes.

Keywords

cardiac complications elective major abdominal surgery perioperative outcomes

Key Takeaways

• The development of cardiac complications following elective major abdominal surgery is associated with increased odds of mortality, perioperative complications, and resource utilization.

• Hospital elective major abdominal surgery and percutaneous coronary intervention volume inversely correlate with the risk of perioperative cardiac complications in this patient population.

Introduction

Major abdominal surgery (MAS) represents a substantial public health burden, with approximately two million procedures performed annually in the US.¹ Despite advancements, approximately 3% of patients develop cardiac complications (CC) following these elective procedures, which are associated with high mortality, perioperative complications, and resource utilization. For instance, prior work from our group has found CC to be associated with up to a 24-fold increase in risk of mortality among patients undergoing MAS.² Furthermore, the development of CC has been linked to increased patient costs often driven by prolonged intensive care unit utilization, hemodynamic monitoring needs, and higher threshold for post-acute care transitions.³

Prior literature has identified several patient and facility characteristics linked with the development of CC following MAS. Established patient-level predictors include advanced age, minority race, and a high burden of comorbidities. Additionally, racial disparities have been documented in broader surgical cohorts. For example, Black race has been independently associated with higher odds of perioperative cardiac arrest (CAR) and failure-to-rescue (FTR) following major elective operations.⁴ Concurrently, institutional factors, particularly hospital operative volume and teaching status, have been shown to influence the recognition and management of these complications. In their analysis of the 2008-2014 National Inpatient Sample (NIS), Sanaiha and colleagues identified hospital operative volume as a critical determinant of survival, attributing superior outcomes at high-volume centers to improved rates of FTR rather than a lower incidence of complications.² However, these prior center-level analyses relied on dated data, leaving a critical gap in our understanding of contemporary trends in perioperative management and outcomes. Moreover, the influence of a hospital’s percutaneous coronary intervention (PCI) volume on outcomes following MAS remains inadequately characterized.⁵

Therefore, in the present study, we evaluated the association of CC with clinical and financial outcomes in a contemporary national cohort of patients undergoing elective MAS. We further sought to characterize patient and hospital factors, including center MAS and PCI volume, associated with the development of CC and mortality. We hypothesized the development of CC to be associated with inferior outcomes. We further hypothesized that increasing institutional MAS and PCI volume would each be independently associated with a reduced risk of CC and FTR.

Methods

This was a retrospective cohort study using the 2016-2022 NIS. As part of the Healthcare Costs and Utilization Project (HCUP), the NIS is the largest all-payer inpatient data repository in the United States. Moreover, the NIS enables the accurate estimation of approximately 97% of all hospitalizations through survey weighting methodology. To account for the complex stratified cluster sampling design of the NIS, all analyses incorporated the NIS-provided discharge weights, designated stratum, and cluster variables, as appropriate, in accordance with HCUP guidelines.

All adult (≥18 years) hospitalizations entailing elective MAS at facilities with at least one incidence of CC were identified using relevant International Classification of Diseases, 10th Revision (ICD-10) procedure and diagnostic codes, respectively.² Colectomy, esophagectomy, gastrectomy, hepatectomy, nephrectomy, pancreatectomy, and splenectomy were considered MAS. Cohorts were stratified by the development of CC, including acute myocardial infarction (AMI) and CAR (CC; Others: non-CC).² Records missing key variables, including age, sex, race, and mortality, were excluded from analysis. Patients admitted to a hospital with MAS volume below the 5th percentile (≤35 cases/year) were also excluded to minimize the effect of outliers (Supplemental Figure 1).

Patient factors, including age, sex, and insurance status, and hospital characteristics, such as bed size and teaching status, were defined in accordance with the NIS data dictionary.⁶ The van Walraven modification of the Elixhauser Comorbidity Index, a validated composite of over 30 conditions, was utilized to quantify the burden of chronic disease.⁷ All mentions of the Elixhauser within this manuscript refer to the van Walraven-modified Elixhauser Comorbidity Index. Specific comorbidities and perioperative complications were defined using previously validated ICD-10 codes.⁸ Complications were categorized into infectious, neurologic (stroke, transient ischemic attack), and respiratory causes (pneumonia, pneumothorax, acute respiratory distress, respiratory failure, prolonged ventilation). Facility annual volume was defined as the total number of cases across all seven included procedures performed at each institution within a calendar year, with each hospital-year as the unit of analysis. Hospitalization costs were derived from total charges using the center-specific cost-to-charge ratios and adjusted for inflation using the 2022 Personal Health Index. The primary endpoint was the development of CC and FTR. Specifically, FTR was defined as mortality following the development of CC. We also examined outcomes associated with CC, including in-hospital mortality, perioperative complications, and resource utilization (hospitalization costs, length of stay (LOS), and nonhome discharge).

Categorical and continuous variables are reported as percentages (%) and medians with interquartile range (IQR), respectively. Unadjusted comparisons were made using Pearson’s Chi-Squared and Wilcoxon rank sum test, as appropriate. Multivariable regression models were developed to evaluate factors associated with the development of CC and FTR. To account for significant intergroup differences in baseline characteristics, entropy balancing was used. Briefly, entropy balancing is a robust re-weighting method to achieve comparable groups while retaining the original sample size and has been shown to be superior to propensity-matching.⁹ Following entropy balancing, additional risk-adjusted analyses using multivariable logistic and linear regression models were developed to assess the independent associations between CC and outcomes. Covariates used in the risk-adjusted analyses include patient demographics, hospital characteristics, comorbidities, surgical modality, and operation type, as appropriate. Restricted cubic splines were used to evaluate a possible non-linear relationship between hospital MAS volume and rate of CC. In brief, this analysis fits cubic polynomial functions to datapoints to estimate non-linear relationships.¹⁰ Additionally, the interaction of institutional PCI intervention, defined as stent placement, and MAS volumes was evaluated using marginal estimates and visualized with a contour plot. Regression outputs are reported as adjusted odds ratios (AORs) or beta coefficients (β) with 95% confidence intervals (CIs). Statistical significance was set at α = 0.05, and all analyses were completed using Stata 18.0 (StataCorp, College Station, Texas).

Results

Among an estimated 904 270 included records, 16 355 (1.8%) developed CC. Compared to non-CC, CC were older (71 vs 62 years, P < 0.001), less frequently female (36.5 vs 52.1%, P < 0.001), and had a higher burden of comorbidities measured by the Elixhauser Comorbidity Index (median: 5 vs 3, P < 0.001). Patients who developed CC had higher rates of congestive heart failure (30.1 vs 4.9%, P < 0.001), chronic kidney disease (8.8 vs 4.8%, P < 0.001), and peripheral vascular disease (10.6 vs 3.5%, P < 0.001) compared to non-CC. Moreover, CC more commonly underwent gastrectomy (15.5 vs 10.2%, P < 0.001), pancreatectomy (14.3 vs 9.2%, P < 0.001), and splenectomy (7.7 vs 5.2%, P < 0.001; Table 1) compared to others.

Table 1.

Demographic, Clinical, and Hospital Characteristics by Cardiac Complications (CC), Including Acute Myocardial Infarction and Cardiac Arrest (Others: Non-CC)

	CC (n = 16,355)	Non-CC (n = 887,915)	P-value
Age (years, median [IQR])	71 [63-77]	62 [52-71]	<0.001
Female	36.5	52.1	<0.001
Elixhauser (median [IQR])	5 [4-7]	3 [1-4]	<0.001
Race			0.001
White	77.0	75.7
Black	11.5	10.3
Hispanic	6.8	8.0
Asian	2.4	2.9
Other	2.4	3.0
Income quartile			<0.001
76-100^th	22.8	26.3
51-75^th	25.3	25.9
26-50^th	26.0	24.5
0-25^th	25.9	23.3
Insurance status			<0.001
Private insurance	22.4	43.4
Medicare	68.6	44.1
Medicaid	6.0	8.2
Other payer	3.0	4.3
Hospital bed size			<0.001
Large	64.0	70.8
Medium	24.9	21.7
Small	11.2	7.5
Teaching status			<0.001
Non-metropolitan	3.8	1.8
Metropolitan nonteaching	13.9	9.4
Metropolitan teaching	82.2	88.8
Hospital region			0.005
Northeast	19.5	21.2
Midwest	24.5	22.7
South	37.2	38.5
West	18.8	17.6
Comorbidities
Cancer	61.4	48.4	<0.001
Congestive heart failure	30.1	4.9	<0.001
Chronic kidney disease	8.8	4.8	<0.001
Diabetes	24.4	20.1	<0.001
History of myocardial infarction	10.7	3.6	<0.001
Peripheral vascular disease	10.6	3.5	<0.001
Operation type
Colectomy	56.0	63.1	<0.001
Esophagectomy	5.5	2.6	<0.001
Gastrectomy	15.5	10.2	<0.001
Hepatectomy	7.2	7.7	0.325
Nephrectomy	10.7	12.0	0.025
Pancreatectomy	14.3	9.2	<0.001
Splenectomy	7.7	5.2	<0.001

CHF, congestive heart failure; CKD, chronic kidney disease; CPD, chronic pulmonary disease; PVD, peripheral vascular disease; MI, acute myocardial infarction; CAR, cardiac arrest.

Following risk-adjustment, older age (AOR 1.02, 95% CI: 1.02-1.03), higher burden of comorbidities (Elixhauser Comorbidity Index: AOR 1.62, 95% CI: 1.59-1.66), and Black race (AOR 1.35, 95% CI: 1.18-1.53) were associated with greater odds of CC. Conversely, female sex (AOR 0.52, 95% CI: 0.48-0.57) and treatment at a metropolitan nonteaching (AOR 0.53, 95% CI: 0.42-0.67) or teaching hospital (AOR 0.28, 95% CI: 0.23-0.35, ref: Non-Metropolitan) were associated with reduced odds of CC (Table 2). Factors associated with FTR following CC include the lowest income quartile (AOR 1.88, 95% CI: 1.42-2.47) and Medicare insurance (AOR 1.30, 95% CI: 1.01-1.68), while Black race (AOR 1.00, 95% CI: 0.75-1.34) did not have a significant relationship. Each 100-case increment in annual institutional MAS volume was independently associated with reduced odds of CC (AOR 0.95, 95% CI: 0.94-0.96, P < 0.001). Likewise, each 100-case increment of annual PCI volume was associated with decreased CC (AOR 0.91, 95% CI: 0.86-0.97, P = 0.003). Specifically, using restricted cubic spline analysis, the inflection point was visualized at 500 MAS cases per year (Figure 1).

Table 2.

Factors Independently Associated With Perioperative Cardiac Complications and Failure to Rescue After the Development of Cardiac Complications

	Cardiac complications		Failure to rescue
	AOR	95% CI	AOR	95% CI
Age	1.02	1.02-1.03*	1.01	0.99-1.02
Female	0.52	0.48-0.57*	1.10	0.92-1.32
Elixhauser (median [IQR])	1.62	1.59-1.66*	1.02	0.96-1.08
Race
White	Reference		Reference
Black	1.35	1.18-1.53*	1.00	0.75-1.34
Hispanic	1.08	0.92-1.27	1.14	0.80-1.61
Asian	0.88	0.68-1.15	1.37	0.80-2.36
Income quartile
76-100^th	Reference		Reference
51-75^th	1.00	0.89-1.12	1.29	0.99-1.67
26-50^th	1.04	0.93-1.17	1.44	1.10-1.89*
0-25^th	1.02	0.90-1.15	1.88	1.42-2.47*
Insurance status
Private insurance	Reference		Reference
Medicare	1.02	0.91-1.15	1.30	1.01-1.68*
Medicaid	1.13	0.94-1.35	1.40	0.94-2.07
Other payer	0.99	0.78-1.26	1.37	0.82-2.29
Hospital bed size
Large	Reference		Reference
Medium	1.54	1.40-1.69*	1.25	1.01-1.55*
Small	2.33	2.05-2.65*	1.09	0.82-1.44
Teaching status
Non-metropolitan	Reference		Reference
Metropolitan nonteaching	0.53	0.42-0.67*	1.09	0.69-1.74
Metropolitan teaching	0.28	0.23-0.35*	0.85	0.56-1.30
Hospital region
Northeast	Reference		Reference
Midwest	0.95	0.84-1.07	0.91	0.68-1.21
South	0.97	0.86-1.09	1.26	0.96-1.64
West	1.13	0.99-1.29	1.08	0.80-1.45

AOR. adjusted odds ratio; CHF, congestive heart failure; CI, confidence interval; CKD, chronic kidney disease; CPD, chronic pulmonary disease; PVD, peripheral vascular disease.

*P < 0.05.

Figure 1.

(A) Risk-adjusted relationship between annual hospital major abdominal surgery (MAS) volume and cardiac complications (CC). The reference line indicates an inflection point at a caseload of 500 per year. (B) Predicted probability of composite CC, including acute myocardial infarction and cardiac arrest, by annual center volume of major abdominal surgery and percutaneous coronary intervention (PCI) admissions in the 2019 National Inpatient Sample. A single year was selected to avoid conflating temporal trends across years. Specifically, year 2019 was selected as a representative pre-pandemic year to avoid COVID-19-related volume disruptions

Patients who developed CC had higher unadjusted in-hospital mortality (27.3 vs 0.5%, P < 0.001) along with infectious (23.9 vs 4.1%, P < 0.001), neurological (1.9 vs 0.2%, P < 0.001), and respiratory complications (41.2 vs 4.3%, P < 0.001) compared to others. Furthermore, relative to non-CC, CC had greater LOS (9 vs 4 days, P < 0.001), hospitalization costs ($39,900 vs $20,400, P < 0.001), and non-home discharge (50.0 vs 6.1%, P < 0.001; Table 3).

Table 3.

Unadjusted and Risk-Adjusted Outcomes Following the Development of Any Cardiac Complications (CC)

	CC	Non-CC	P-value	AOR/β	95% CI
In-hospital mortality (%)	27.3	0.5	<0.001	10.53	8.90-12.46
Complications (%)
Infectious	23.9	4.1	<0.001	1.98	1.76-2.23
Neurological	1.9	0.2	<0.001	1.30	0.90-1.89
Respiratory	41.2	4.3	<0.001	3.53	3.18-3.91
Resource utilization
LOS (days [IQR]/β)	9 [5-15]	4 [3-6]	<0.001	1.90	1.62-2.19
Cost ($1000s [IQR]/β)	39.9 [24.0-65.0]	20.4 [14.4-28.0]	<0.001	8.51	7.36-9.65
Nonhome discharge (%)	50.0	6.1	<0.001	3.71	3.36-4.10

*AOR, adjusted odds ratio; IQR, interquartile range; LOS, length of stay.

Following entropy-balancing and risk adjustment, CC remained associated with increased odds of mortality (AOR 10.53, 95% CI: 8.90-12.46), infectious (AOR 1.98, 95% CI: 1.76-2.23), and respiratory complications (AOR 3.53, 95% CI: 3.18-3.91). Furthermore, the development of CC was linked with greater LOS (β 1.90 days, 95% CI: 1.62-2.19), inpatient hospitalization costs (β $8,500, 95% CI: $7400-$9700), and odds of non-home discharge (AOR 3.71, 95% CI: 3.36-4.10; Table 3).

Upon subgroup analysis with cohorts stratified by AMI, those who developed AMI were older (71 vs 63 years, P < 0.001), less frequently female (39.7 vs 51.9%, P < 0.001), and had a higher comorbidity burden as measured by the Elixhauser Comorbidity Index (median: 5 vs 3, P < 0.001) compared to others. Patients who developed AMI had higher rates of congestive heart failure (37.5 vs 5.2%, P < 0.001), chronic kidney disease (8.5 vs 4.8%, P < 0.001), and peripheral vascular disease (10.1 vs 3.6%, P < 0.001). Moreover, AMI patients more commonly underwent gastrectomy (15.3 vs 10.3%, P < 0.001), pancreatectomy (12.9 vs 9.3%, P < 0.001), and splenectomy (8.7 vs 5.2%, P < 0.001; Supplementary Table 2).

Patients who developed AMI had higher unadjusted in-hospital mortality (18.2 vs 0.8%, P < 0.001) along with infectious (28.4 vs 4.3%, P < 0.001), neurological (2.8 vs 0.2%, P < 0.001), and respiratory complications (47.3 vs 4.7%, P < 0.001) compared to others. Furthermore, relative to non-AMI patients, AMI had with greater LOS (10 vs 4 days, P < 0.001), hospitalization costs ($47,500 vs $20,500, P < 0.001), and non-home discharge (48.5 vs 6.6%, P < 0.001).

Following entropy-balancing and risk adjustment, AMI remained associated with increased odds of mortality (AOR 1.28, 95% CI: 1.05-1.57), infectious (AOR 1.66, 95% CI: 1.37-2.00), and respiratory complications (AOR 2.32, 95% CI: 1.97-2.75). Furthermore, the development of AMI was linked with greater LOS (β 1.76 days, 95% CI: 1.22-2.29), inpatient hospitalization costs (β $9,900, 95% CI: $7900-$11,900), and odds of non-home discharge (AOR 1.63, 95% CI: 1.38-1.92). Finally, each 100-case increment in annual institutional MAS was independently associated with reduced odds of AMI (AOR 0.97, 95% CI: 0.96-0.99, P = 0.02). Likewise, each 100-case increment of annual PCI volume was associated with decreased AMI (AOR 0.72, 95% CI: 0.51-0.99, P = 0.04).

Similarly, subgroup analyses stratified by CAR revealed that compared to non-CAR, those who developed CAR were older (71 vs 63 years, P < 0.001), less frequently female (35.2 vs 52.1%, P < 0.001), and had a higher comorbidity burden as measured by the Elixhauser Comorbidity Index (median: 5 vs 3, P < 0.001). Patients who developed CAR also had higher rates of congestive heart failure (28.9 vs 5.0%, P < 0.001), chronic kidney disease (8.7 vs 4.8%, P < 0.001), and peripheral vascular disease (10.7 vs 3.5%, P < 0.001). Moreover, CAR patients more commonly underwent gastrectomy (15.5 vs 10.2%, P < 0.001), pancreatectomy (14.8 vs 9.2%, P < 0.001), and splenectomy (7.5 vs 5.2%, P < 0.001; Supplementary Table 3).

Patients who developed CAR had higher unadjusted in-hospital mortality (27.2 vs 0.5%, P < 0.001) along with infectious (23.7 vs 4.1%, P < 0.001), neurological (1.8 vs 0.2%, P < 0.001), and respiratory complications (41.8 vs 4.3%, P < 0.001) compared to others. Furthermore, relative to non-CAR patients, CAR was associated with greater LOS (9 vs 4 days, P < 0.001), hospitalization costs ($39,600 vs $20,400, P < 0.001), and non-home discharge (49.4 vs 6.1%, P < 0.001).

Following entropy-balancing and risk adjustment, CAR remained associated with increased odds of mortality (AOR 6.86, 95% CI: 5.88-8.01), infectious (AOR 1.64, 95% CI: 1.45-1.85), and respiratory complications (AOR 2.76, 95% CI: 2.48-3.07). Furthermore, the development of CAR was linked with greater LOS (β 1.69 days, 95% CI: 1.38-1.99), inpatient hospitalization costs (β $6,900, 95% CI: $5600-$8100), and odds of non-home discharge (AOR 3.08, 95% CI: 2.77-3.41). Finally, each 100-case increment increase in annual institutional MAS volume was independently associated with reduced odds of CAR (AOR 0.92, 95% CI: 0.91-0.94, P < 0.001). Likewise, each 100-case increment increase in annual PCI volume was associated with decreased odds of CAR (AOR 0.92, 95% CI: 0.87-0.98, P = 0.01).

Discussion

In this national analysis, CC following elective MAS was associated with markedly increased mortality and resource utilization. Furthermore, we identified numerous patient factors to be associated with the risk of CC and associated FTR. Interestingly, increasing institutional volume of MAS and PCI were independently linked with reduced risk of CC. With implications for guidelines, risk-stratification, and hospital accreditation, these findings warrant further discussion.

Congruent with prior literature, CC was linked with increased inpatient mortality, perioperative complications, and resource utilization.^2,11 The observed increase in risk of mortality is likely attributable to ischemia that leads to a rapid cascade of multi-organ failure.¹² From a financial standpoint, these worsened clinical outcomes translate into substantial cost burdens for the health system. Historically, administrators have hesitated to fund comprehensive preoperative cardiac testing or prehabilitation, arguing low yield.⁵ The substantial hospitalization costs associated with CC highlighted in this study underscore the economic burden of these events and may motivate future cost-effectiveness analyses of prevention strategies. Additionally, recent evidence-based guidelines have advocated for perioperative biomarker surveillance (eg, BNP/troponin). While our study does not evaluate these strategies directly, the substantial mortality and cost burden we report provide contextual support for the economic rationale underlying such investment.³ Our findings suggest that greater efforts are justified toward reducing cardiac causes of complications following MAS.

We further identified Black race as an important predictor of CC, consistent with prior reports.⁴ This suggests that structural differences in health are not fully captured by standard metrics like the Elixhauser Comorbidity Index. In contrast, FTR was driven not by race, but by economic factors where patients in the lowest income bracket had higher odds compared to the wealthiest quartile. This disparity likely stems from low-income patients being more frequently funneled into safety-net hospitals or under-resourced centers.¹³ Indeed, hospitals with lower FTR rates tend to have higher hospitalist, overnight physician, and dedicated rapid team coverage.¹⁴ Consequently, despite best efforts, these resource constraints create a critical gap in rescue capabilities following complications. Furthermore, current risk calculators underestimate risk for these populations by failing to account for these non-clinical factors.¹⁵ For example, the revised cardiac risk index, commonly used for perioperative cardiovascular risk assessment, focuses exclusively on clinical factors and fails to account for social determinants of health.¹⁶ Limited considerations for these influences may in part be due to the absence of comprehensive measures in contemporary data sets. Future policy and guidelines should consider including variables associated with structural determinants of outcomes for more comprehensive risk assessment for surgical candidates.

Finally, we found that increasing institutional volume for MAS was independently associated with improved outcomes. Our study showing positive correlation between surgical volume and superior outcomes supports policy initiatives like the Leapfrog Group’s referral standards.¹⁷ However, while the Leapfrog Group has set volume thresholds for specific operations like pancreatectomy and esophagectomy, most abdominal procedures lack the strong evidence to justify regionalization policies. This can be explained by qualitative research suggesting that outcomes are driven by multifaceted hospital, surgeon, and team-level mechanisms beyond institutional volume alone.¹⁸ Interestingly, our finding that hospital PCI volume independently predicts outcomes for MAS patients identifies broader hospital capabilities as another indicator. High-volume PCI centers offer unique resources, such as 24/7 STEMI teams, whereas others may lack the immediate interventional availability required to manage acute ischemia.¹⁹ Given these risks, timely access to cardiac intervention is critical for patients experiencing CC following MAS. Currently, the American College of Surgeons recommends hospitals have interventional cardiac potential or formalized, rapid transfer protocols to partner with capable centers for emergency general surgery. Our findings prompt the expanded consideration of these processes for elective MAS, especially in high-risk patients.⁵ Collectively, these volume-outcome relationships support centralization policies. However, centralization policies carry important second-order consequences, including disparate impacts on rural and underserved populations with limited travel capacity. Furthermore, patient preference data suggest that many individuals are willing to accept a modestly higher complication risk in exchange for care at a local rather than a distant center, which remains an important consideration that must be weighed alongside volume-outcome data in any policy discussion. As such, centralization policy implementation must be balanced against access considerations.

This study has several important limitations inherent to the use of administrative databases. As a retrospective analysis of the NIS, our analyses are susceptible to variations in practice, potential coding errors, and lack of sensitivity for subtle complications given its reliance on ICD-10 codes.⁶ As well, we were unable to explicitly identify the precise timing of complications relative to the index operation. Moreover, the data set lacks granular clinical information such as laboratory values, intraoperative hemodynamics, or angiographic findings, limiting our ability to confirm the specific etiology of cardiac events. Our analysis of hospital characteristics was also limited by the lack of data on surgeon-specific experience or specific ICU staffing models, which are known to influence rescue rates. Furthermore, we restricted our analysis to hospitals reporting at least one CC event, which may have led to exclusion of facilities with true zero event rates and could affect the estimated inflection point in our volume-outcome analysis. Moreover, although surgical modality was included as a covariate in risk-adjusted analyses, this study did not include an additional subgroup analysis by operative approach. Finally, as the NIS does not track patients across calendar years, our ability to assess long-term survival or readmissions beyond the index year was restricted. However, while there are limitations to NIS, this database captures ∼97% of the US population to present generalizable findings. We also employed multiple approaches to statistical analyses to conduct a robust investigation.

In summary, this study demonstrates that the development of CC after MAS is associated with significant mortality and financial burden, driven by a complex interplay of clinical, socioeconomic, and institutional factors. Additionally, increasing institutional MAS and PCI volume were independently associated with improved outcomes. Future research should focus on integrating social determinants into preoperative risk algorithms and evaluating center capabilities in addition to volume thresholds to determine center excellence. Ultimately, improving outcomes requires a paradigm shift, more comprehensive risk evaluations, and systemic changes, to better predict CC.

Supplemental Material

Supplemental material - Factors and Outcomes Associated With Cardiac Complications Among Patients Undergoing Elective Major Abdominal Surgery

Supplemental material for Factors and Outcomes Associated With Cardiac Complications Among Patients Undergoing Elective Major Abdominal Surgery by Zihan Gao, MHSc, Troy N. Coaston, MSCR, Preston Leung, MDiv, Jeffrey Balian, BS, Giselle Porter, BS, Areti Tillou, MD, Yas Sanaiha, MD, and Peyman Benharash, MD in The American Surgeon™

Footnotes

ORCID iDs

Zihan Gao

Peyman Benharash

Ethical Considerations

Given deidentified nature of these data, this study was exempt from full review by the Institutional Review Board at the University of California, Los Angeles.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Author PB receives proctor fees from AtriCure as a surgical proctor. This submission does not analyze or use products or services from AtriCure. All other authors declare no conflict of interest.

Presentation

This manuscript was presented at the 2026 American College of Surgeons Southern California Chapter Annual Scientific Meeting.

Supplemental Material

Supplemental material for this article is available online.

References

Carney

Weissler

Fox

Tecce

Hsu

Fischer

. Trends in open abdominal surgery in the united states—observations from 9,950,759 discharges using the 2009–2013 national inpatient sample (NIS) datasets. Am J Surg. 2017;214(2):287-292. doi:10.1016/j.amjsurg.2017.01.001.

Sanaiha

Juo

Aguayo

, et al. Incidence and trends of cardiac complications in major abdominal surgery. Surgery. 2018;164(3):539-545. doi:10.1016/j.surg.2018.04.030.

Farooq

Malik

Bhat

Farooq

. Perioperative cardiac complications and evidence-based strategies for their management. Cureus. 2025;17(10):e95276. doi:10.7759/cureus.95276.

Williamson

Sanaiha

Tran

, et al. Disparities in cardiac arrest and failure to rescue after major elective noncardiac operations. Surgery. 2022;171(5):1358-1364. doi:10.1016/j.surg.2021.09.001.

Thompson

Fleischmann

Smilowitz

, et al. AHA/ACC/ACS/ASNC/HRS/SCA/SCCT/SCMR/SVM guideline for perioperative cardiovascular management for noncardiac surgery: a report of the American college of cardiology/american heart association joint committee on clinical practice guidelines. Circulation. 2024;150(19):e351-e442. doi:10.1161/CIR.0000000000001285.

HCUP-US NIS Overview. Accessed July 3, 2025.https://hcup-us.ahrq.gov/nisoverview.jsp

van Walraven

Austin

Jennings

Quan

Forster

. A modification of the elixhauser comorbidity measures into a point system for hospital death using administrative data. Med Care. 2009;47(6):626-633. doi:10.1097/MLR.0b013e31819432e5.

Coaston

Vadlakonda

Mallick

, et al. Decade-long variation in the development of complications after cardiac surgery across the United States. Surgery. 2025;183:109387. doi:10.1016/j.surg.2025.109387.

Hainmueller

. Entropy balancing for causal effects: a multivariate reweighting method to produce balanced samples in observational studies. Polit Anal. 2012;20(1):25-46. doi:10.1093/pan/mpr025.

10.

Gauthier

Gooley

. Cubic splines to model relationships between continuous variables and outcomes: a guide for clinicians. Bone Marrow Transplant. 2020;55(4):675-680. doi:10.1038/s41409-019-0679-x.

11.

STARSurg collaborative, EuroSurg collaborative . Impact of postoperative cardiovascular complications on 30-day mortality after major abdominal surgery: an international prospective cohort study. Anaesthesia. 2024;79(7):715-724. doi:10.1111/anae.16220.

12.

Ruetzler

Smilowitz

Berger

, et al. Diagnosis and management of patients with myocardial injury after noncardiac surgery: a scientific statement from the American heart association. Circulation. 2021;144(19):e287-e305. doi:10.1161/CIR.0000000000001024.

13.

Popescu

Fingar

Cutler

Guo

Jiang

. Comparison of 3 safety-net hospital definitions and association with hospital characteristics. JAMA Netw Open. 2019;2(8):e198577. doi:10.1001/jamanetworkopen.2019.8577.

14.

Ward

Dimick

Zhang

Campbell

Ghaferi

. Association between hospital staffing models and failure to rescue. Ann Surg. 2019;270(1):91-94. doi:10.1097/SLA.0000000000002744.

15.

Khan

Coresh

Pencina

, et al. Novel prediction equations for absolute risk assessment of total cardiovascular disease incorporating cardiovascular-kidney-metabolic health: a scientific statement from the American heart association. Circulation. 2023;148(24):1982-2004. doi:10.1161/CIR.0000000000001191.

16.

Fleisher

Beckman

Brown

, et al. ACC/AHA 2007 guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: a report of the American college of cardiology/american heart association task force on practice guidelines (writing committee to revise the 2002 guidelines on perioperative cardiovascular evaluation for noncardiac surgery) developed in collaboration with the American society of Echocardiography, American society of nuclear cardiology, heart rhythm society, society of cardiovascular anesthesiologists, society for cardiovascular angiography and interventions, society for vascular medicine and biology, and society for vascular surgery. J Am Coll Cardiol. 2007;50(17):e159-e241. doi:10.1016/j.jacc.2007.09.003.

17.

Birkmeyer

Dimick

. Potential benefits of the new leapfrog standards: effect of process and outcomes measures. Surgery. 2004;135(6):569-575. doi:10.1016/j.surg.2004.03.004.

18.

Mesman

Faber

Westert

Berden

HJJM

. Dutch surgeons’ views on the volume-outcome mechanism in surgery: A qualitative interview study. Int J Qual Health Care. 2017;29(6):797-802. doi:10.1093/intqhc/mzx100.

19.

Jacobs

Ali

Best

, et al. Systems of care for ST-Segment–Elevation myocardial infarction: a policy statement from the American heart association. Circulation. 2021;144(20):e310-e327. doi:10.1161/CIR.0000000000001025.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.21 MB