Differences Between Ischemic and Nonischemic Cardiomyopathy in Heart Failure Related Cardiogenic Shock.
Feb 26, 2025DOI: 10.1016/j.cardfail.2025.01.020Journal ArticleJCF
Written by: Jason FeinmanMatthew I TomeyMichael G PalazzoloMiguel MartilloMaria RonquilloNoah MossGregory SerraoErin A BohulaDavid D BergSean VAN DiepenJason N KatzMeshe D ChondeSunit-Preet ChaudhryAlvin J GeorgeDaniel GerberMichael J GoldfarbNorma M KellerMichael C KontosDaniel B LoriauxConnor G O'BrienBarbara A PisaniAlastair ProudfootKiran SidhuShashank S SinhaLakshmi SridharanNatalie TapaskarAlexander ThomasAndrea D ThompsonDavid A MorrowUmesh GidwaniEvan Leibner
Find these authors on Roon
SC
Dr. Sunit-Preet Chaudhry
Advanced Heart Failure · Ascension St. Vincent Hospital
CO
Dr. Connor O'Brien
Cardiac Critical Care · UCSF Medical Center
NT
Dr. Natalie Tapaskar
Advanced Heart Failure · UT Southwestern
EL
Dr. Evan Leibner
Critical Care · Mount Sinai
Introduction
Contemporary registries have demonstrated that heart failure-related cardiogenic shock (HF-CS), in the absence of an acute myocardial infarction (AMI), accounts for a majority of CS admissions to cardiac intensive care units (CICUs), and mortality rates as high as 30%.1 Historically, patients with HF have had their cardiomyopathy categorized into ischemic (ICM) and nonischemic (NICM) cohorts, with conflicting data concerning the relevance of this categorization to prognosis.2 The implications of this characterization of cardiomyopathy on outcomes of HF-CS are incompletely understood. We sought to compare patients’ characteristics, treatment modalities and clinical outcomes of HF-CS in patients with HF attributed to ICM and NICM.
Methods
The Critical Care Cardiology Trials Network (CCCTN) registry is a collaborative registry of consecutive patients admitted to CICUs across North America and is overseen by the Thrombolysis in Myocardial Infarction (TIMI) Study Group (Boston, MA). The registry’s methods have been published.3 Participating centers collect data for a minimum annual 2-month collection cycle. Admissions to participating centers between 2017 and 2022 were included in this analysis. The institutional review board at each participating center approved the CCCTN registry protocol and waiver of informed consent.Patients with CS were identified by using a standard definition of systolic blood pressure < 90 mmHg for > 30 minutes (or requiring vasopressor support) along with evidence of end-organ dysfunction.1 Shock etiology was classified by the site investigator as cardiogenic, distributive, hypovolemic, or mixed. Cardiogenic shock was then further subclassified into AMI-CS, secondary CS (eg, cardiac tamponade or valvular heart disease) or HF-CS, which is the focus of this analysis. Site investigators at each participating center independently categorized the etiology of cardiomyopathy based on all available data at the time of data capture.
Results
Among the 2463 admissions of people with HF-CS included in the analysis, 902 (36.6%) had ICM, and 1561 (63.4%) had NICM (Table 1). Compared to those admitted with NICM, patients admitted with ICM were older and more commonly male and Caucasian and more likely to have major comorbidities, including diabetes mellitus and chronic kidney disease. Rates of previously diagnosed HF were similar, but there were fewer cases with left ventricular ejection fraction ≥ 50% in the ICM cohort.
Patients admitted with ICM more often suffered a cardiac arrest prior to CICU presentation (21.8% vs 12.2%; P < 0.001). Peak Society for Cardiovascular Angiography & Interventions (SCAI) shock stages were similar, but Sequential Organ Failure Assessment (SOFA) scores were higher in patients admitted with ICM.
Invasive hemodynamic data were available for 1586 (64.4%) admissions. Invasive hemodynamic monitoring was less common in ICM admissions (60.3% vs 66.8%; P = 0.001). The rates of mechanical circulatory support (MCS), vasopressor or inotrope usage at the time of hemodynamic assessment were similar (74% vs 71%; P = 0.25). Admissions with ICM exhibited slightly higher cardiac indexes (2.0 vs 1.9 l/min/m2; P < 0.001) and lower central venous pressures (14 vs 15 mmHg; P = 0.007) but without differences in pulmonary capillary wedge pressure (24.5 vs 24.0mmHg; P = 0.49). Right ventricular and biventricular failure was less commonly seen in presentations with ICM compared to those with NICM. For example, the right atrial to pulmonary capillary wedge ratio was lower in ICM compared to NICM (0.5 vs 0.6; P < 0.001), and the pulmonary artery pulsatility index was significantly higher in those with ICM (1.9 vs 1.6; P < 0.001).
In-hospital treatments are displayed in Fig. 1. People admitted with ICM were more likely to receive mechanical ventilation and renal replacement therapies. The usage rates of intravenous inotropes or any temporary MCS device were similar in the cohorts.
In-hospital outcomes are summarized in Fig. 2, A. In-hospital mortality rates were significantly higher among admissions with ICM (32.5% vs 21.8 %; P < 0.001). People with NICM were more likely to undergo cardiac transplantation during their hospitalization, but the rates of durable MCS did not differ. Accordingly, the proportion of those who survived to discharge without cardiac transplantation or durable LVAD placement did not differ between groups.
After controlling for differences in age, sex, diabetes mellitus, chronic kidney disease, shock severity, and preceding cardiac arrest, ICM was independently associated with a higher rate of in-hospital mortality (OR 1.56, 95% CI 1.26–1.93; P < 0.001) (Fig. 2, B).The higher in-hospital mortality rate associated with ICM was maintained across key subgroups, including de novo HF, age, sex, and those who received invasive hemodynamic monitoring or temporary MCS usage (Fig. 2, B).
Discussion
In this analysis of CICU admissions of people with HF-CS, ICM as the etiology of HF-CS portended a worse prognosis than NICM. Patients with ICM had substantially more comorbid conditions, which likely contributed to their less favorable outcomes; however, the higher risk of patients with ICM remained evident after adjusting for key differences in comorbidities, shock severity and higher rates of preceding cardiac arrest.
The patterns of hemodynamics, as well as site-adjudicated descriptions, in this analysis reflected higher rates of right ventricular and biventricular dysfunction in the NICM cohort. These parameters might suggest worse outcomes for NICM, because both biventricular congestion and elevated central venous pressure have been associated with worsened outcomes in all-comers with cardiogenic shock.4 In our study, this association appears to be overshadowed by the etiology of HF and other associated comorbidities.
Despite these differences in comorbidities and hemodynamics, the management of patients with ICM and NICM was similar. There were no differences in inotrope usage or overall temporary MCS. This pattern was consistent for both intra-aortic balloon pumps (IABPs) as well as for advanced MCS. Prior research has suggested improved hemodynamic impact with IABPs if the cardiogenic shock was due to HF rather than AMI.5 Whether this relationship extends further to differences between ischemic and nonischemic etiologies of HF-CS is a potential area of future study. In our analysis, the point estimates for mortality continued to favor NICM vs ICM, whether or not MCS was used as a part of the management strategy. The differences in mechanical ventilation and renal-replacement therapies are likely to be influenced by the differences in baseline comorbidities and preceding cardiac arrest.
The association between NICM and higher rates of transplantation is consistent with prior research.6 The referral rates for advanced therapies and individual-center rationales are not captured within this registry, but it is possible that the lower rate of heart transplantation among HF-CS patients with ICM may, in part, be due to the advanced ages and higher rates of baseline comorbidities or to social factors, such as tobacco usage or multiorgan dysfunction in this cohort.7 The rates of durable MCS device placement, in contrast, were not different between groups. Right and biventricular dysfunction was more prevalent in the NICM cohort, and it implies less favorable candidacy for durable univentricular support, but the higher rates of comorbidities in the ICM cohort may have reduced the perceived benefit and referrals in this cohort as well, leading to net lower transplantation rates. Both durable MCS and cardiac transplantation are potential options for appropriate patients with advanced HF, and they have been demonstrated to improve outcomes.2,8 Whether this increased usage of advanced therapies in NICM explains the observed lower mortality rates and the reasons for this usage discrepancy may be areas of future investigation.
Conclusion
In this contemporary multicenter registry of CICUs across North America, patients with HF-CS had higher in-hospital mortality rates if the etiology of their cardiomyopathy was due to a chronic ischemic as compared to a nonischemic cause.
EL
Dr. Evan Leibner
Emergency Medicine · Critical CareMount Sinai Health SystemevanlNew York City, NY, USA
