Risk Factors and Outcomes of Protein-Calorie Malnutrition in Chronic Heart Failure Patients Undergoing Elective Cardiac Surgery

Introduction: Chronic heart failure (CHF) patients are often malnourished. Our aim was to determine the effect of protein-calorie malnutrition (PCM) on in-hospital outcomes in CHF patients following elective cardiac surgery and to identify risk factors for PCM in this patient population. Methods: A retrospective analysis of the National Inpatient Sample (NIS) database was conducted from 2016 to 2018. In-hospital outcomes in adult patients with CHF undergoing elective coronary artery bypass graft (CABG) with cardiopulmonary bypass-assist or cardiac valve replacement surgeries were analyzed. Propensity-score matching was used to match CHF patients with and without PCM and followed by logistic regression analysis. A multivariate logistic regression model was used to identify the risk factors associated with PCM in this population. Results: In total 25,940 CHF patients were identified, of which 6,271 underwent elective CABG and 19,669 underwent valve replacement surgeries. The prevalence of PCM in CHF patients undergoing CABG and valve replacement was 3.9% and 2.9%, respectively. CHF patients with PCM had significantly higher risk of in-hospital mortality, post-operative cardiac and gastrointestinal complications compared to CHF patients without PCM. The mean hospital length of stay was twice as high in the PCM group (mean days: 18.6 vs 9.9). Female gender, Black race (vs White race), a high Charlson Comorbidity Index, Medicare/Medicaid insurance status (vs private insurance), and CHF (systolic and combined systolic and diastolic) were independently associated with significantly higher risk of PCM diagnosis. Conclusions: CHF patients with PCM who undergo elective CABG or valve replacement surgeries are at a significantly higher risk of mortality, post-operative cardiac and gastrointestinal complications, and increased duration of hospital stay compared to those without PCM. Future prospective studies should assess the CHF patients who are at a higher risk of PCM and whether correcting pre-operative nutrition in this surgical population can improve outcomes following cardiac surgery.


Introduction
Among patients undergoing cardiac surgery, the prevalence of malnutrition may be 18-40% at the time of surgery [1][2][3]. Protein-calorie malnutrition (PCM) develops from insufficient protein and/or caloric (i.e., energy) intake. This can result in unintended weight loss, including loss of muscle mass or loss of subcutaneous fat, tissue fluid accumulation (e.g., low oncotic pressures), and diminished physical function [4]. In addition, PCM has significant impact on multiple systems of the body including weakness of skeletal muscles, reduced heart muscle mass, impaired wound healing, immune deficiency, and hypothermia [5]. Previous research has indicated that PCM leads to an increased risk of morbidity and mortality, hospital length of stay (LOS), and complications in cardiac surgery patients [6]. Current data on in-hospital outcomes for chronic heart failure (CHF) patients with PCM undergoing cardiac surgery is limited.
In the presence of CHF, the intestine and the heart develop a bidirectional relationship, the clinical manifestation of which is cardiointestinal syndrome (CIS) [7]. Accordingly, advanced heart failure (HF) is commonly associated with gastrointestinal (GI) symptoms, including abdominal pain, anorexia, and proteinlosing enteropathy. Conditions that affect both these systems, such as PCM, stand out as under-explored comorbidities that may affect cardiac surgery outcomes. The systemic inflammation and stress of cardiac surgery can affect PCM patients more profoundly than their well-nourished counterparts whose morbidity and mortality after surgery are significantly lower [8]. The objective of this retrospective cohort analysis was to investigate in-hospital outcomes of CHF patients with and without PCM following elective cardiac surgery. Additionally, we sought to identify risk factors associated with PCM in patients undergoing elective cardiac surgery. This article was previously presented as a meeting abstract at the 2022 Society of Cardiac Anesthesiology Meeting on May 15, 2022.

Data collection
A retrospective population-based analysis utilizing 2016-2018 National Inpatient Sample (NIS) data was performed [9]. From 2016 onward, the NIS data uses the International Classification of Diseases 10th edition, Clinical Modification/Procedure Coding System (ICD-10-CM/PCS). The ICD-10-CM/PCM codes were used to conduct the analysis. The NIS is one of several nationwide databases part of the Healthcare Cost and Utilization Project (HCUP), Agency for Healthcare Research and Quality. The NIS contains de-identified patient information and conforms to Health Insurance Portability and Accountability Act (HIPAA) guidelines. Hence, written informed consent was not required per the exemption by the institutional review board (IRB). The authors adhered to the HCUP Data Use Agreement. Strengthening the reporting of observational studies in epidemiology (STROBE) checklist for cohort studies was followed.

Study cohort
Adult patients (age ≥18 years) that underwent coronary artery bypass graft (CABG) surgery with cardiopulmonary bypass-assist (CPB) or valve replacement surgeries (i.e., aortic, pulmonic, mitral, or tricuspid) during elective hospital admission were included. Patients undergoing elective CABGs or valve replacement surgeries were categorized into CHF and no heart failure (HF) groups. If patients did not have an acute or chronic HF diagnosis code, they were categorized as having no HF. Additionally, among the CHF group, those with a diagnosis of PCM were identified using the Centers for Medicare and Medicaid Services (CMS) definition. The CMS has defined PCM as any two of the following: (1) muscle wasting and loss of subcutaneous fat, (2) <50-75% of normal nutritional intake over >1-2 weeks, (3) reduced functional capacity, and (4) weight loss [11].

Outcomes
The primary outcome was the risk of in-hospital mortality in CHF patients with pre-existing PCM vs no PCM following elective cardiac surgery (i.e., CABG or valve replacement), while secondary outcomes included hospital LOS and the risk of acute cardiac and GI complications. The composite cardiac complications included -acute MI, cardiogenic shock, cardiac arrest, ventricular tachycardia (VT)/ventricular fibrillation (VF), cardiac tamponade, and intra-aortic balloon pump (IABP) use. The composite GI complications included -GI bleeding, intestinal ischemia or infarction, and acute hepatic failure. Additional secondary outcomes included assessing the risk factors for a PCM diagnosis in patients undergoing cardiac surgery to identify vulnerable populations.

Statistical analysis
For assessing differences in categorical variables a Chi-square test was used, while for continuous variables a two-sample t-test was used for assessing the mean difference. Data are presented as frequency (%) or mean (standard deviation {SD}) for categorical and continuous variables, respectively. In the assessment of the primary and secondary outcomes between PCM and non-PCM CHF groups, a 1:6 propensity-score matching (PSM) was utilized to minimize the probability of selection bias. Matching factors included demographic characteristics (e.g., age, gender, race), comorbidities (i.e., CCI), and insurance status. Following PSM, logistic regression was used to assess the effects of PCM in CHF patients undergoing cardiac surgery on the primary and secondary outcomes. The categorical outcomes are reported as odds ratio (OR) with a 95% confidence interval (CI) and a p-value (significant if <0.05). Finally, for the identification of potential risk factors for PCM diagnosis in patients undergoing elective cardiac surgery, a multivariate logistic regression model was utilized to calculate the OR. Statistical software for data sciences, STATA 14 (College Station, TX: StataCorp LLC), was used for data extraction and analysis.

Discussion
In CHF patients undergoing elective cardiac surgeries, pre-existing PCM is associated with significantly higher inpatient mortality and post-operative cardiac (i.e., cardiac arrest, cardiogenic shock, VT/VF, cardiac tamponade, use of IABP) and GI (i.e., GI bleeding, intestinal ischemia or infarction, and acute hepatic failure) complications. These findings support previously reported poorer outcomes (e.g., post-operative complications, mortality) in patients with PCM undergoing esophageal [12] and lung surgery [13], percutaneous coronary intervention [14], and transcatheter aortic valve replacement procedures [15,16]. In our cohort, patients with CHF (vs no HF) were more likely to be suffering from PCM. In addition, in CHF patients, PCM was associated with twice as longer hospitalization compared to patients without PCM. This is consistent with our current understanding of the relationship between PCM and multiple organ systems, and the downstream cascade induced by cardiac surgery and cardiopulmonary bypass.
PCM is a complex disease process involving nearly every organ system in the body. The hallmark of PCM is inadequate protein or energy intake, which leads to a chronic systemic inflammatory state termed cytokineinduced malnutrition [5]. The downstream effect of this chronic inflammatory state is a negative feedback loop, which further exacerbates the problem. Sarcopenia develops as the body adapts to this chronic state of malnutrition by reducing the basal metabolic rate and shifting muscle protein from cardiac and respiratory muscles to central lean tissues (liver, GI tract, renal, hematologic, and immune systems) [5]. Cardiac cachexia can ensue following the development of advancing HF and wasting syndrome may be present in over 20% of hospitalized HF patients [17]. In this study, CHF patients undergoing cardiac surgeries were twice as likely to have PCM compared to those without HF. Impairments of systolic cardiac function tend to have higher prevalence of nutritional derangements, although patients with diastolic dysfunction can also develop malnutrition, which is consistent with our findings [18].
The multisystem nature of PCM and resulting pathophysiology, in the context of CHF, contribute to increased mortality, complications, and hospital LOS for cardiac surgery patients [17,19,20]. GI hypoperfusion and bowel wall edema from inadequate splanchnic perfusion in HF predispose to bacterial translocation and endotoxemia, triggering release of cytokines and other pro-inflammatory mediators [7,21]. These circulating cytokines exert direct toxic effects on the heart, leading to depressed myocardial contractility and cardiogenic shock (and increased use of IABP), as well as an increased incidence of arrhythmias that can result in cardiac arrest. Hence, cardiointestinal syndrome (CIS) from decompensated HF leads to a bidirectional positive feedback loop that perpetuates cardiac impairment and worsens PCM. Furthermore, the increased gut permeability and right-sided filling pressures (e.g., high venous pressures) from advancing HF commonly lead to GI tract endothelial dysfunction, intestinal edema and ischemia/infarction, hepatic congestion and dysfunction (e.g., sinusoidal edema that limits oxygen diffusion to hepatocytes), resulting in an increased incidence of GI complications [7]. This is consistent with our findings that showed CHF patients who had PCM before undergoing cardiac surgery had higher risk of cardiac and GI complications and had longer hospital LOS compared to CHF patients without PCM. The increased long-term mortality observed in hospitalized HF patients with malnutrition may be partially attributed to the increase in cardiac and GI complications as observed in our study population [19,20].
Cardiac surgery and CPB can further exacerbate the aforementioned derangements [16]. Surgical trauma activates neutrophils, endothelial cells, platelets, and pro-inflammatory mediators, while CPB initiates an inflammatory response and ischemia-reperfusion injury [22]. Ultimately, this results in increased oxidative stress and systemic inflammation, leading to acute and persistent organ injury and dysfunction. As this study showed CHF patients with PCM undergoing cardiac surgery are susceptible to a higher incidence of complications and prolonged hospital stays.
This study also showed that certain subgroups of patients appear to have a higher risk of PCM diagnosis preoperatively. For instance, compared to those with private insurance, Medicare and Medicaid patients have a higher incidence of PCM. Among other factors, this might be due to poor dietary intake in this population. Females also show a higher incidence of PCM diagnosis. Smaller body habitus and decreased protein reserve likely predispose females to PCM compared to their male counterparts. Systolic and combined (systolic and diastolic) HF patients are nearly twice as likely to be diagnosed with PCM. This finding is likely the result of the body's adaptation to inadequate protein intake. An intra-cellular shift of muscle proteins from the cardiac and respiratory systems to the liver, GI tract, kidney, hematologic, and immune systems results in cardiac cachexia. Cardiac cachexia is more detrimental to the patient with systolic failure (or combined systolic and diastolic) than diastolic failure alone, as the failing heart is left unable to adequately contract. Lastly, increasing age places patients at increased risk of PCM [23]. These patients not only have a reduction in lean tissue mass and metabolic reserves but also decreased organ function and ability to tolerate oxidative stress [23]. Interestingly, elderly patients undergoing only valve surgery were not associated with an increased risk for PCM, which may be due to the increased therapeutic use of transcatheter approaches in this patient population.
Increased comorbid conditions, substantial surgical insult, and use of CPB make the post-operative cardiac patient vulnerable to complications. Pre-operative identification and correction of PCM in this patient population is prudent. The Malnutrition Universal Screening Tool has been shown to predict mortality in cardiovascular surgery patients. The use of such screening tools should be assessed in CHF patients undergoing cardiac surgery in future studies to identify patients who are malnourished or at risk [1]. It has been demonstrated that intense nutritional support is beneficial in certain subsets of patients at high risk for PCM [24]. In fact, the European Society for Clinical Nutrition and Metabolism guidelines recommend medical nutrition therapy for seven to 14 days pre-operatively for severely malnourished patients, even if this delays surgery [25]. Compared to CHF patients with PCM, those without PCM are likely to have decreased cytokine-induced malnutrition, thereby attenuating the inflammatory response and downstream effects. Optimizing metabolic reserves prior to operation likely confers a similar benefit. Physicians, including anesthesiologists involved in the peri-operative assessment and optimization of patients for cardiac surgery, should focus on the early identification and treatment of PCM utilizing the expertise of a registered dietitian.
This study is limited by its retrospective design and reliance upon a documented diagnosis of PCM. PCM may therefore be underdiagnosed or underreported. In addition, multiple scoring systems exist for identification of PCM, and there is no standardized definition of PCM diagnosis. Hence, the generalizability of these results is challenging across this patient population and across centers in the United States. Additionally, due to the retrospective nature and the use of a non-granular database, the inclusion of a variety of diseasespecific variables, and other intra-operative factors that could impact the outcomes of this study are omitted. Moreover, retrospective studies that examine associations based on administrative datasets are hypothesis-generating rather than proof of relationships. Hence, prospective studies are needed to confirm the findings of this study.

Conclusions
CHF patients undergoing elective CABG or valve replacement surgeries are twice as likely to have preexisting PCM compared to patients without HF. PCM is more likely to be present in patients with systolic or combined CHF (systolic and diastolic CHF) patients. Following elective cardiac surgery, CHF patients with PCM (vs without PCM) have higher hospital mortality, post-operative cardiac and GI complications, and longer duration of hospital stay. Nutritional status in this surgical population should be assessed using standardized definitions by prospective studies to determine whether timely correction pre-operatively improves prognosis. This appears most critical for individuals who are females, Black, with multiple comorbidities, on governmental insurance plans such as Medicare/Medicaid, and with systolic CHF. Hence, future prospective studies on CHF patients undergoing cardiac surgeries need to identify the type of nutritional problem and assess whether implementation of a personalized dietary plan prior to cardiac surgery can improve outcomes.

Additional Information Disclosures
Human subjects: All authors have confirmed that this study did not involve human participants or tissue. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.