Author's summary
This study addresses the unmet need for clarity on sex differences in the management and outcomes of cardiogenic shock (CS) patients with ischemic cardiomyopathy (ICMP) and non-ICMP etiologies. Their multicenter analysis of 1,247 patients from the RESCUE registry revealed no significant sex-based disparities in in-hospital management, 30-day, or 12-month mortality outcomes, even after propensity-score matching. These findings suggest that sex does not independently influence prognosis in CS, underscoring the importance of equitable care. This study bridges gaps in understanding the role of sex in CS outcomes and highlights avenues for future research in personalized treatment strategies.
Keywords: Shock, cardiogenic; Sex characteristics; Etiology; Cardiomyopathies; Prognosis
Abstract
Background and Objectives
Comprehensive data on sex-based differences in the management and outcomes of patients with and without ischemic cardiomyopathy (ICMP) presenting with cardiogenic shock (CS) remain limited. This study aimed to investigate whether clinical management and outcomes differ by sex among CS patients, stratified by underlying etiology.
Methods
We analyzed 1,247 CS patients from the RESCUE registry, a multicenter observational cohort, stratified by sex and CS etiology: ICMP (females: 276, males: 730) and non-ICMP (females: 111, males: 130). Primary outcomes included in-hospital and 12-month mortality. Multivariable Cox proportional hazards models and propensity-score matching were used to adjust for confounding factors.
Results
Among ICMP patients, females were less likely to undergo coronary angiography (p=0.001), although rates of successful revascularization were similar between sexes (p=0.982). In-hospital 30-day mortality did not differ significantly between females and males in either the ICMP cohort (37.1% vs. 29.5%; adjusted hazard ratio [HR], 0.93; 95% confidence interval [CI], 0.63–1.39; p=0.737) or the non-ICMP cohort (28.3% vs. 25.6%; adjusted HR, 1.23; 95% CI, 0.68–2.22; p=0.493). At 12 months, mortality risk remained comparable between sexes in both ICMP (46.4% vs. 37.1%; adjusted HR, 0.82; 95% CI, 0.57–1.17; p=0.274) and non-ICMP groups (40.1% vs. 41.3%; adjusted HR, 0.91; 95% CI, 0.56–1.45; p=0.685). These findings were consistent after propensity-score matching.
Conclusions
There was no significant difference in management, 12-month or in-hospital mortality between females and males, irrespective of the etiology of CS.
Trial Registration
ClinicalTrials.gov Identifier: NCT02985008
Graphical Abstract
INTRODUCTION
Cardiogenic shock (CS) represents the most severe form of acute heart failure and is characterized by life-threatening tissue hypoperfusion due to inadequate cardiac output.1) Previous research has highlighted sex disparities in the in-hospital management of CS and their adverse impacts on clinical outcomes, particularly among female patients with ischemic cardiomyopathy (ICMP).2),3) Efforts have been made to address these disparities, yet recent studies have often included heterogeneous populations, resulting in conflicting findings. For instance, a retrospective single-center cohort study of overall CS patients found no significant sex-based differences in in-hospital management or survival rates.4) Conversely, a national registry study focusing on CS complicating acute myocardial infarction (MI) reported that females were less likely to receive guideline-recommended care, leading to poorer in-hospital outcomes.3) Only one study has specifically evaluated sex differences in CS patients stratified by etiology (ICMP and non-ICMP), and it reported notable sex disparities in management and outcomes regardless of etiology. However, that study was limited to in-hospital outcomes and did not explore the relationship between sex disparities in management and long-term clinical outcomes.2) In this context, the current study aimed to investigate the associations between sex and both in-hospital and 12-month clinical outcomes according to CS etiology using a dedicated, large-scale, multicenter CS registry.
METHODS
Ethical statement
The study protocol was approved by the Samsung Medical Center Ethics Committee (approval number 2016-03-130, April 6, 2016) and the ethics committees of all participating centers. The study was conducted in accordance with the principles of the Declaration of Helsinki (2013 revision). Written informed consent was obtained from all prospectively enrolled patients prior to participation. For retrospectively enrolled patients, the requirement for informed consent was waived by the Institutional Review Boards of the respective participating hospitals.
Study population
The study population was drawn from the REtrospective and prospective observational Study to investigate Clinical oUtcomes and Efficacy of left ventricular assist device for Korean patients with cardiogenic shock (RESCUE) registry (ClinicalTrials.gov NCT02985008, Registered December 5, 2016), a multicenter database encompassing both retrospective and prospective data on patients with CS.5) Between January 2014 and December 2018, a total of 1,247 consecutive patients aged over 19 years with CS (954 enrolled retrospectively and 293 enrolled prospectively) were enrolled from 12 tertiary centers in the Republic of Korea. Inclusion criteria were (1) older than 19 years, (2) systolic blood pressure lower than 90 mmHg for at least 30 minutes or requirement for inotropic or vasopressor support to achieve a systolic blood pressure higher than 90 mmHg, and (3) pulmonary congestion with signs of impaired organ perfusion, including altered mental status, cold peripheries, urine output less than 0.5 mL/kg/hr over the previous 6 hours or blood lactate level exceeding 2.0 mmol/L. Exclusion criteria were (1) out-of-hospital cardiac arrest, (2) alternative causes of shock, and (3) refusal of active treatment. The 1,247 patients were categorized into 2 groups based on the etiology of CS: 1,006 patients (80.7%) were assigned to the ICMP group and 241 patients (19.3%) to the non-ICMP group. Patients were classified into the ICMP group if CS was primarily attributed to an ischemic cause. A history of previous MI or percutaneous coronary intervention (PCI) alone was not sufficient to classify a patient as having ICMP. Patients with prior MI or PCI were included in the non-ICMP group if coronary angiography (CAG) did not reveal a significant ischemia-producing lesion responsible for the shock, and if another etiology (e.g., myocarditis, stress-induced cardiomyopathy, or tachycardia-induced cardiomyopathy) was suspected. We assessed in-hospital mortality and 12-month clinical outcomes based on sex (Supplementary Figure 1).
Data collection and study outcomes
Clinical patient demographics, in-hospital management, laboratory data, procedural data, and outcome data were collected by independent clinical research coordinators using web-based case report forms. All baseline data were recorded upon patient admission. Additional information was obtained from medical records or via telephone contact when necessary.
Primary outcomes were 30-day in-hospital mortality and 12-month mortality. Secondary outcomes included recurrent MI, unplanned repeat revascularization, readmission due to heart failure, and cerebrovascular accidents during the 12-month follow-up period. All-cause mortality was defined as death from any cause, while cardiac mortality was defined as death due to a cardiac condition. Recurrent MI was identified by the recurrence of symptoms or electrocardiographic changes associated with an increase in a cardiac biomarker above the upper limit of normal. Periprocedural MI was not included as a clinical outcome. Clinically driven revascularization occurring after discharge from the index hospitalization was classified as an unplanned repeat revascularization event. Since recurrent MI and unplanned repeat revascularization were not considered relevant clinical outcomes in the non-ICMP group, these parameters were analyzed only in the ICMP cohort. The vasoactive-inotropic score (VIS) was calculated using the maximal administration rates of dopamine, dobutamine, epinephrine, milrinone, vasopressin, and norepinephrine during the first 48 hours of shock.6)
The RESCUE registry defines outcome criteria in accordance with the parameters outlined by The Academic Research Consortium-2 (ARC-2) consensus. Consequently, the definitions of clinical events and outcomes in this study were based on the ARC-2 consensus.7) Analyses were limited to 12 months of follow-up due to variations in follow-up duration. Median follow-up duration was 308 days (interquartile range: 12–373) for the ICMP cohort and 341 days (interquartile range: 18–393) for the non-ICMP cohort.
Statistical analysis
Discrete and categorical variables are presented as numbers and relative frequency (percentages). Continuous variables were compared using Student’s t-test and are reported as mean ± standard deviation. Categorical data were analyzed using the χ2 test and are presented as count and relative frequency. The Kaplan-Meier method was used to estimate the cumulative incidence of clinical events, with comparisons using the log-rank test. Hazard ratio (HR) with 95% confidence interval (CI) for primary and secondary outcomes were estimated using Cox proportional hazards models. The proportionality assumption was evaluated using Schoenfeld residuals and graphically through log-log plots.
A multivariable Cox model was constructed using variables with p value <0.10 from univariable analyses, as well as clinically relevant variables. For the overall population, the final model for the ICMP cohort included age, body mass index (BMI), hypertension, diabetes mellitus, current smoker, previous MI, hemoglobin, total bilirubin, N-terminal pro B-type natriuretic peptide (NT-proBNP), and CAG. For the non-ICMP cohort, the model contained age, diabetes mellitus, current smoker, previous PCI, hemoglobin, and peak troponin-I. Among patients who survived to discharge, the final model for the ICMP group incorporated age, BMI, hypertension, current smoker, previous MI, heart rate, hemoglobin, total bilirubin, peak creatine kinase-myocardial band, NT-proBNP, CAG, and length of hospital stay. For the non-ICMP group, the model included age, current smoker, systolic blood pressure, hemoglobin, and peak troponin-I.
For the sensitivity analysis, propensity-score matching was conducted using the propensity scores derived from a multivariable logistic regression model. The balance between the 2 groups after matching was assessed by calculating standardized mean differences, which were approximately 10% for most matched covariates, indicating successful balance. A subgroup analysis of the primary outcome was performed based on clinical and procedural factors of interest, comparing females and males within both the ICMP and non-ICMP groups.
To assess the normality of continuous variables, we performed the Kolmogorov-Smirnov test. When the sample size was sufficiently large (n≥30), normality was assumed based on the central limit theorem. For continuous variables that followed a normal distribution, Student’s t-test was used, and results were presented as mean ± standard deviation. For variables that did not follow a normal distribution, as determined by the Kolmogorov-Smirnov test, the Wilcoxon rank-sum test was performed, and results were reported as median (25th percentile to 75th percentile).
All probability values were 2-sided, with p values <0.05 considered statistically significant. Statistical analyses were performed using SPSS version 27.0 (SPSS Inc., Chicago, IL, USA) and R version 4.4.0 (R Foundation for Statistical Computing, Vienna, Austria).
RESULTS
Baseline clinical characteristics and in-hospital management
Baseline clinical characteristics and in-hospital management characteristics of the ICMP group are summarized in Table 1. Among the ICMP cohort, 276 patients (27.4%) were female, and 730 patients (72.6%) were male. Female patients were older, smoked less frequently, and had a lower BMI than their male counterparts. Hypertension was more prevalent among female patients. NT-proBNP levels were significantly higher in females than males, while hemoglobin and total bilirubin levels were significantly lower. Notably, female patients underwent CAG less frequently than males (88.4% vs. 95.9%; p<0.001), although the rates of successful revascularization were similar between the 2 groups (86.9% vs. 86.6%; p=0.982). Use of extracorporeal membrane oxygenation (ECMO), extracorporeal cardiopulmonary resuscitation (ECPR), and intra-aortic balloon pump (IABP) was comparable between females and males. Other in-hospital management factors, such as shock to ECMO time, duration of ECMO support, mechanical ventilation, renal replacement therapy, VIS, and length of intensive care unit or total hospital stay did not differ significantly between the 2 sexes in the ICMP group (Table 1). In the propensity score-matched cohort, there were no significant differences in baseline clinical characteristics or in-hospital management between female and male patients (Supplementary Table 1).
Table 1. Baseline clinical characteristics and in-hospital management in the ischemic cardiomyopathy group.
| Total (n=1,006) | Female (n=276) | Male (n=730) | p value | |||
|---|---|---|---|---|---|---|
| Baseline clinical characteristics | ||||||
| Age (years) | 67.2±12.4 | 73.2±11.1 | 64.9±12.0 | <0.001 | ||
| Body mass index (kg/m2) | 23.6±3.4 | 22.9±3.7 | 23.9±3.3 | <0.001 | ||
| Hemodynamics | ||||||
| Systolic blood pressure (mmHg) | 74.8±30.2 | 75.9±32.1 | 74.3±29.4 | 0.452 | ||
| Diastolic blood pressure (mmHg) | 47.2±20.3 | 46.6±20.5 | 47.4±20.3 | 0.573 | ||
| Heart rate (beat/min) | 80.3±32.9 | 79.9±33.7 | 80.4±32.6 | 0.812 | ||
| Medical history | ||||||
| Hypertension | 568 (56.5) | 188 (68.1) | 380 (52.1) | <0.001 | ||
| Diabetes mellitus | 378 (37.6) | 112 (40.6) | 266 (36.4) | 0.255 | ||
| Dyslipidemia | 293 (29.1) | 76 (27.5) | 217 (29.7) | 0.546 | ||
| Chronic kidney disease | 95 (9.4) | 29 (10.5) | 66 (9.0) | 0.556 | ||
| Current smoker | 321 (31.9) | 18 (6.5) | 303 (41.5) | <0.001 | ||
| Previous PCI | 154 (15.3) | 35 (12.7) | 119 (16.3) | 0.185 | ||
| Previous myocardial infarction | 141 (14.0) | 30 (10.9) | 111 (15.2) | 0.096 | ||
| Peripheral artery disease | 39 (3.9) | 8 (2.9) | 31 (4.2) | 0.421 | ||
| Previous history of stroke | 91 (9.0) | 27 (9.8) | 64 (8.8) | 0.706 | ||
| Left ventricular ejection fraction (%) | 37.6±15.8 | 38.2±15.3 | 37.3±16.0 | 0.478 | ||
| Laboratory findings | ||||||
| Hemoglobin (g/dL) | 12.8±2.5 | 11.3±2.1 | 13.4±2.4 | <0.001 | ||
| Creatinine (serum) (mg/dL) | 1.5±1.5 | 1.5±1.3 | 1.6±1.5 | 0.470 | ||
| Glucose (mg/dL) | 227.6±118.3 | 233.7±137.4 | 225.3±110.4 | 0.375 | ||
| Total bilirubin (mg/dL) | 0.8±1.1 | 0.7±0.6 | 0.9±1.2 | 0.006 | ||
| Lactic acid (mmol/L) | 6.7±4.6 | 6.6±4.6 | 6.7±4.5 | 0.718 | ||
| Peak troponin-I (ng/mL) | 65.6±148.4 | 64.7±154.9 | 65.9±146.0 | 0.908 | ||
| Peak CK-MB (ng/mL) | 207.8±398.9 | 181.2±363.0 | 217.7±411.2 | 0.175 | ||
| NT-proBNP (pg/mL) | 7,824.1±12,248.3 | 12,037.6±14,555.0 | 6,119.2±10,741.0 | <0.001 | ||
| In-hospital managements | ||||||
| In-hospital procedures | ||||||
| Coronary angiography | 944 (93.8) | 244 (88.4) | 700 (95.9) | <0.001 | ||
| Successful revascularization | 817 (86.6) | 212 (86.9) | 605 (86.6) | 0.982 | ||
| ECMO | 360 (35.8) | 90 (32.6) | 270 (37.0) | 0.223 | ||
| ECPR | 200 (19.9) | 51 (18.5) | 149 (20.4) | 0.551 | ||
| Shock to ECMO time (minute) | 337.8±750.2 | 339.1±709.4 | 337.4±765.0 | 0.985 | ||
| ECMO duration (day) | 4.9±4.9 | 4.9±5.7 | 4.9±4.6 | 0.990 | ||
| Intra-aortic balloon pump | 298 (29.6) | 81 (29.3) | 217 (29.7) | 0.968 | ||
| Mechanical ventilation | 559 (55.6) | 152 (55.1) | 407 (55.8) | 0.902 | ||
| Renal replacement therapy | 209 (20.8) | 48 (17.4) | 161 (22.1) | 0.124 | ||
| Vasoactive-inotropic score | 73.9±139.9 | 81.7±160.7 | 70.9±131.2 | 0.321 | ||
| Length of ICU stay (day) | 9.7±21.3 | 10.4±23.2 | 9.5±20.6 | 0.579 | ||
| Length of hospital stay (day) | 16.0±25.6 | 17.7±26.2 | 15.4±25.4 | 0.210 | ||
Data are presented as mean ± standard deviation or number (%).
CK-MB = creatine kinase-myocardial band; ECMO = extracorporeal membrane oxygenation; ECPR = extracorporeal cardiopulmonary resuscitation; ICU = intensive care unit; NT-proBNP = N-terminal pro B-type natriuretic peptide; PCI = percutaneous coronary intervention.
Table 2 presents the baseline clinical characteristics and in-hospital management characteristics of the non-ICMP group. In this non-ICMP cohort, 111 patients (46.1%) were female and 130 (53.9%) were male. Dilated cardiomyopathy was the most common cause of CS in both sexes (27.0% in females and 35.4% in males). Female patients smoked less frequently than male patients. Serum hemoglobin levels were lower and peak troponin-I levels were higher in female than male patients. Other baseline characteristics in the non-ICMP group were not significantly different between the sexes. In terms of in-hospital management, there were no significant differences between female and male patients, including use of ECMO, ECPR, and IABP, as well as shock-to-ECMO time, duration of ECMO support, mechanical ventilation, and renal replacement therapy (Table 2). After propensity-score matching, no significant differences in baseline clinical characteristics or in-hospital management between female and male patients were observed (Supplementary Table 2).
Table 2. Baseline clinical characteristics and in-hospital management in the non-ischemic cardiomyopathy group.
| Total (n=241) | Female (n=111) | Male (n=130) | p value | |||
|---|---|---|---|---|---|---|
| Baseline clinical characteristics | ||||||
| Age (years) | 59.0±17.3 | 59.9±17.9 | 58.2±16.9 | 0.448 | ||
| Body mass index (kg/m2) | 22.6±3.9 | 22.6±4.2 | 22.5±3.5 | 0.897 | ||
| Hemodynamics | ||||||
| Systolic blood pressure (mmHg) | 68.0±23.5 | 66.7±20.9 | 69.1±25.5 | 0.416 | ||
| Diastolic blood pressure (mmHg) | 44.0±18.0 | 43.1±16.2 | 44.8±19.5 | 0.457 | ||
| Heart rate (beat/min) | 93.4±37.6 | 91.4±35.3 | 95.2±39.5 | 0.449 | ||
| Cause of shock | 0.392 | |||||
| Dilated cardiomyopathy | 76 (31.5) | 30 (27.0) | 46 (35.4) | |||
| Fulminant myocarditis | 40 (16.6) | 23 (20.7) | 17 (13.1) | |||
| Valvular heart disease | 19 (7.9) | 11 (9.9) | 8 (6.2) | |||
| Refractory arrhythmia | 31 (12.9) | 12 (10.8) | 19 (14.6) | |||
| Pulmonary thromboembolism | 24 (10.0) | 11 (9.9) | 13 (10.0) | |||
| Other causes | 51 (21.2) | 24 (21.6) | 27 (20.8) | |||
| Medical history | ||||||
| Hypertension | 92 (38.2) | 43 (38.7) | 49 (37.7) | 0.973 | ||
| Diabetes mellitus | 65 (27.0) | 24 (21.6) | 41 (31.5) | 0.113 | ||
| Dyslipidemia | 37 (15.4) | 14 (12.6) | 23 (17.7) | 0.362 | ||
| Chronic kidney disease | 28 (11.6) | 16 (14.4) | 12 (9.2) | 0.294 | ||
| Current smoker | 35 (14.5) | 4 (3.6) | 31 (23.8) | <0.001 | ||
| Previous PCI | 21 (8.7) | 5 (4.5) | 16 (12.3) | 0.056 | ||
| Previous myocardial infarction | 19 (7.9) | 5 (4.5) | 14 (10.8) | 0.119 | ||
| Peripheral artery disease | 13 (5.4) | 5 (4.5) | 8 (6.2) | 0.780 | ||
| Previous history of stroke | 28 (11.6) | 13 (11.7) | 15 (11.5) | 1.000 | ||
| Left ventricular ejection fraction (%) | 35.0±18.2 | 37.2±18.7 | 33.3±17.7 | 0.130 | ||
| Laboratory findings | ||||||
| Hemoglobin (g/dL) | 12.1±2.9 | 11.1±2.4 | 12.9±3.0 | <0.001 | ||
| Creatinine (serum) (mg/dL) | 1.6±1.0 | 1.5±1.1 | 1.7±1.0 | 0.207 | ||
| Glucose (mg/dL) | 201.7±119.7 | 212.2±141.7 | 192.9±97.5 | 0.245 | ||
| Total bilirubin (mg/dL) | 1.8±3.2 | 1.7±4.3 | 1.8±1.9 | 0.778 | ||
| Lactic acid (mmol/L) | 6.5±4.5 | 6.6±4.1 | 6.5±4.8 | 0.801 | ||
| Peak troponin-I (ng/mL) | 15.4±50.0 | 24.6±69.9 | 7.6±19.0 | 0.016 | ||
| Peak CK-MB (ng/mL) | 71.2±140.9 | 74.3±136.8 | 68.6±144.7 | 0.757 | ||
| NT-proBNP (pg/mL) | 12,220.9±12,622.8 | 13,407.7±13,255.7 | 11,248.5±12,073.6 | 0.297 | ||
| In-hospital managements | ||||||
| In-hospital procedures | ||||||
| Coronary angiography | 71 (29.5) | 27 (24.3) | 44 (33.8) | 0.140 | ||
| Successful revascularization | 9 (12.7) | 1 (3.7) | 8 (18.2) | 0.158 | ||
| ECMO | 136 (56.4) | 63 (56.8) | 73 (56.2) | 1.000 | ||
| ECPR | 40 (16.6) | 19 (17.1) | 21 (16.2) | 0.979 | ||
| Shock to ECMO time (minute) | 598.8±1,031.0 | 490.3±787.0 | 692.0±1,199.7 | 0.250 | ||
| ECMO duration (day) | 7.1±7.2 | 6.3±4.4 | 7.7±8.7 | 0.292 | ||
| Intra-aortic balloon pump | 16 (6.6) | 9 (8.1) | 7 (5.4) | 0.557 | ||
| Mechanical ventilation | 150 (62.2) | 71 (64.0) | 79 (60.8) | 0.706 | ||
| Renal replacement therapy | 76 (31.5) | 38 (34.2) | 38 (29.2) | 0.488 | ||
| Vasoactive-inotropic score | 71.4±121.6 | 79.9±131.6 | 64.1±112.3 | 0.318 | ||
| Length of ICU stay (day) | 19.3±30.6 | 16.4±28.2 | 21.8±32.5 | 0.174 | ||
| Length of hospital stay (day) | 32.4±39.5 | 30.7±37.7 | 33.9±41.1 | 0.543 | ||
Data are presented as mean ± standard deviation or number (%).
CK-MB = creatine kinase-myocardial band; ECMO = extracorporeal membrane oxygenation; ECPR = extracorporeal cardiopulmonary resuscitation; ICU = intensive care unit; NT-proBNP = N-terminal pro B-type natriuretic peptide; PCI = percutaneous coronary intervention.
Clinical outcomes
In-hospital outcomes
Rates of in-hospital outcomes are presented in Supplementary Table 3. In the ICMP group, female patients had higher rates of in-hospital mortality (40.2% vs. 31.4%; p=0.010) and in-hospital cardiac mortality (35.1% vs. 26.9%; p=0.012) than male patients. Percentage of patients receiving hemodynamic support via ECMO was similar between females and males (32.6% vs. 37.0%; p=0.223). Among ICMP patients who received ECMO support, there were no significant differences in the risks of cannular site bleeding, cerebrovascular accidents, limb ischemia, gastrointestinal bleeding, or sepsis between the sexes.
In the non-ICMP group, the rates of in-hospital mortality (32.4% vs. 33.1%; p=1.000) and in-hospital cardiac mortality (27.9% vs. 26.9%; p=0.976) were similar between female and male patients. ECMO utilization frequency was comparable between the sexes (56.8% vs. 56.2%; p=1.000). However, among non-ICMP patients undergoing ECMO, cannular site bleeding was significantly more frequent in females (23.8% vs. 4.1%; p=0.002), while other ECMO-related complications did not differ significantly between sexes.
Following propensity-score matching, in-hospital mortality (36.9% vs. 36.4%; p=1.000) and in-hospital cardiac mortality (31.8% vs. 32.3%; p=1.000) in the ICMP group were similar between female and male patients. ECMO-related complications did not differ by sex in the ICMP group. In the non-ICMP group, in-hospital mortality (30.4% vs. 35.7%; p=0.688) and in-hospital cardiac mortality (23.2% vs. 28.6%; p=0.666) were not significantly different between sexes. Additionally, there were no significant differences in ECMO-related complications between female and male patients.
Thirty-day in-hospital clinical outcomes
A comparison of 30-day in-hospital clinical outcomes between female and male patients in the ICMP and non-ICMP groups is presented in Figure 1. In the ICMP group, there was no significant difference in the risk of all-cause mortality (37.1% vs. 29.5%; adjusted HR, 0.93; 95% CI, 0.63–1.39; p=0.737) or cardiac mortality (34.0% vs. 26.0%; adjusted HR, 0.90; 95% CI, 0.59–1.38; p=0.641) between female and male patients (Figure 1A and B). Similarly, in the non-ICMP group, the risk of all-cause mortality (28.3% vs. 25.6%; adjusted HR, 1.23; 95% CI, 0.68–2.22; p=0.493) and cardiac mortality (25.9% vs. 22.7%; adjusted HR, 1.18; 95% CI, 0.63–2.22; p=0.595) did not differ significantly between the sexes (Figure 1C and D). In sensitivity analysis with propensity-score matching, there were no significant differences in the risk of all-cause mortality or cardiac mortality between female and male patients, regardless of the etiology of CS (Supplementary Table 4).
Figure 1. Time-to-event Kaplan-Meier survival curves of 30-day cumulative incidence of all-cause mortality and cardiac mortality. (A) Kaplan-Meier curves of all-cause mortality in patients with ICMP. (B) Kaplan-Meier curves of cardiac mortality in patients with ICMP. (C) Kaplan-Meier curves of all-cause mortality in patients with non-ICMP. (D) Kaplan-Meier curves of cardiac mortality in patients with non-ICMP.
CI = confidence interval; HR = hazard ratio; ICMP = ischemic cardiomyopathy.
Twelve-month follow-up clinical outcomes
In the ICMP group, there were no significant differences in the risk of all-cause mortality (46.4% vs. 37.1%; adjusted HR, 0.82; 95% CI, 0.57–1.17; p=0.274) or cardiac mortality (40.5% vs. 32.6%; adjusted HR, 0.84; 95% CI, 0.56–1.23; p=0.365) at 12 months between female and male patients (Figure 2A and B). Additionally, the risks of recurrent MI, unplanned repeat revascularization, readmission due to heart failure, and cerebrovascular accidents at 12 months were similar between the sexes (Supplementary Figure 2). This pattern persisted in the sensitivity analysis using propensity-score matching (Supplementary Table 5, Figure 2).
Figure 2. Time-to-event Kaplan-Meier survival curves of 12-month cumulative incidence of all-cause mortality and cardiac mortality. (A) Kaplan-Meier curves of all-cause mortality in patients with ICMP. (B) Kaplan-Meier curves of cardiac mortality in patients with ICMP. (C) Kaplan-Meier curves of all-cause mortality in patients with non-ICMP. (D) Kaplan-Meier curves of cardiac mortality in patients with non-ICMP.
CI = confidence interval; HR = hazard ratio; ICMP = ischemic cardiomyopathy.
Similarly, in the non-ICMP group, there were no significant differences in the risk of all-cause mortality (40.1% vs. 41.3%; adjusted HR, 0.91; 95% CI, 0.56–1.45; p=0.685) or cardiac mortality (36.2% vs. 34.2%; adjusted HR, 0.94; 95% CI, 0.56–1.57; p=0.806) at 12 months between female and male patients (Figure 2C and D). However, the risk of readmission due to heart failure at 12 months was significantly higher in male patients (p=0.021), while the risk of cerebrovascular accidents at 12 months was comparable between sexes (p=0.411) (Supplementary Figure 3). After propensity-score matching, differences in clinical outcomes at 12 months remained insignificant between sexes, including the risk of readmission due to heart failure (p=0.113) (Supplementary Table 5, Figure 3).
Figure 3. Comparative subgroup HRs for all-cause mortality according to sex. (A) Comparison of all-cause mortality between sexes according to subgroups in the ICMP population. (B) Comparison of all-cause mortality between sexes according to subgroups in the non-ICMP population.
CI = confidence interval; CPR = cardiopulmonary resuscitation; HR = hazard ratio; ICMP = ischemic cardiomyopathy.
Twelve-month follow-up clinical outcomes for patients who survived to discharge
For patients who survived to discharge, there were no significant differences between females and males in the risks of all-cause mortality, cardiac mortality, recurrent MI, unplanned repeat revascularization, readmission due to heart failure, and cerebrovascular accidents at 12 months in the ICMP group. In the non-ICMP group, while the risks of all-cause mortality and cardiac mortality were similar between the sexes, male patients had a significantly higher risk of readmission due to heart failure at 12 months than female patients (Supplementary Table 6). Baseline clinical characteristics and in-hospital management characteristics are presented in Supplementary Table 7.
Subgroup analysis
Subgroup analyses were performed to evaluate the association between sex and all-cause mortality in various contexts. In the ICMP group, the higher risk of 12-month all-cause mortality for female patients was consistent across subgroups, except for the renal replacement therapy subgroup, where females receiving renal replacement therapy showed a lower risk of mortality than their male counterparts (HR, 0.89 vs. 1.80; p=0.003 for interaction). Significant interactions were also observed in the subgroups of mechanical ventilation (HR, 1.28 vs. 2.22; p=0.043 for interaction) and mechanical circulatory support (MCS) (HR, 1.17 vs. 2.27; p=0.007 for interaction) (Figure 3A). In the non-ICMP group, there were no significant differences in the risk of all-cause mortality at 12 months between females and males across subgroups, except the BMI subgroup, where obese males showed a lower risk of mortality than females (HR, 3.00 vs. 0.75; p=0.007 for interaction) (Figure 3B).
DISCUSSION
The current study investigated sex differences in in-hospital and 12-month clinical outcomes among stratified CS patients, categorized by etiology, using a dedicated, large-scale, multicenter real-world CS registry. The main finding was that in-hospital and 12-month mortality in patients with CS did not vary significantly between sexes, regardless of the etiology of CS. In-hospital management was equitable, with no observed sex disparities in treatment approaches and procedures performed. Sensitivity analysis using propensity-score matching confirmed that sex differences were not an independent prognostic factor for in-hospital or 12-month mortality, irrespective of CS etiology. Interestingly, female patients in the non-ICMP group had significantly lower rates of readmission due to heart failure in the multivariable analysis; however, this significance was not upheld in the sensitivity analysis. To the best of our knowledge, this is the first study to evaluate sex disparities according to shock etiology in CS patients with clinical outcomes up to 12 months.
Clinical studies of the effects of sex on management strategies and outcomes of patients with CS have reported heterogeneous results.2),3),8),9) In studies that showed worse clinical outcomes in females, females received significantly less intensive therapy than their male counterparts.2),9) Osman et al.2) and Cenko et al.9) reported significant sex disparities in in-hospital management for patients with ICMP and CS, with females receiving less frequent MCS or revascularization than males, resulting in higher in-hospital or 30-day mortality; however, these studies analyzed only short-term clinical outcomes and lacked analysis of post-discharge follow-up clinical outcomes. Notably, Elgendy et al.3) analyzed both short- and long-term outcomes and reported sex differences in in-hospital outcomes driven by management disparities but equitable 12-month clinical outcomes, suggesting that sex disparities in management primarily influence short-term clinical outcomes. Meanwhile, previous CS studies that reported no significant sex differences in clinical outcomes showed no sex disparities in in-hospital management, such as MCS application, use of inotropes or vasopressors, renal replacement therapy, or mechanical ventilation.4),10) Exceptionally, Yan et al.11) reported similar 30-day mortality between sexes, although female patients were less likely to be treated with percutaneous left ventricular assist devices and more likely to be treated with catecholamines or vasopressors. However, those studies that reported similar clinical outcomes between sexes were performed in patients with heterogeneous CS etiologies and analyzed only in-hospital clinical outcomes. We analyzed management disparities and clinical outcomes over a 12-month follow-up period according to the etiology of CS, as well as the association between sex and all-cause mortality across various contexts in subgroup analyses. Significant interactions were observed depending on whether renal replacement therapy, mechanical ventilation, or MCS was applied, indicating that more intensive therapy could decrease differences in clinical outcomes between sexes. Collectively, we showed that equitable management translated into similar in-hospital clinical outcomes, with consistently insignificant differences in 12-month clinical outcomes between sexes, regardless of CS etiology. Our results can be interpreted in a similar context to previous studies, suggesting a positive correlation between the absence of sex disparities and decreased in-hospital mortality, and vice versa.
Considering the high acuity and critical status of patients with CS, many clinical events, including deaths, occur early in the diagnosis of CS.3) Therefore, research on long-term clinical outcomes in CS patients is scarce, and it is undetermined whether CS patients who survive to discharge from index hospitalization have a similar clinical course as patients with chronic cardiomyopathy according to sex. We additionally analyzed clinical outcomes from discharge up to 12 months for patients who survived to discharge to determine if female patients with CS have better prognoses, similar to those observed in chronic heart failure with ICMP or non-ICMP.12),13) In CS patients with ICMP, there was no significant difference in unadjusted 12-month all-cause mortality between females and males. However, after multivariable adjustment or propensity-score matching, females had a lower risk of mortality than males. Although the event rate was relatively lower than that reported in studies on chronic heart failure patients with ICMP, we observed similar sex differences in post-discharge clinical outcomes in surviving patients to those of patients with chronic heart failure due to ICMP.12) This trend of clinical outcomes similar to those of chronic heart failure patients was also observed in CS patients with non-ICMP who survived to discharge. In our study, male patients had significantly higher risks of 12-month readmission due to heart failure in the unadjusted analysis, and this difference remained significant after multivariable adjustment. These findings suggest that CS females with ICMP or non-ICMP have better clinical outcomes than males, potentially due to estrogen’s cardioprotective effects, similar to its role in chronic heart failure, which requires further investigation.14) Our study highlights the importance of considering sex differences in the management and long-term follow-up of CS patients. Despite the critical nature of CS and the immediate risks associated with it, the long-term outcomes post-discharge appear to reflect patterns seen in chronic heart failure, where females often have better prognoses. This underscores the need for tailored post-discharge care strategies that account for these differences and further research to understand the underlying mechanisms driving these outcomes.
Interestingly, in the non-ICMP group, subgroup analysis demonstrated that among patients with higher BMI, males had a significantly lower risk of 12-month all-cause mortality than their female counterparts. Moreover, obese males had a lower risk of mortality than non-obese males, while obese females had a higher mortality rate than non-obese females. Previous studies have shown that the obesity paradox, which is the paradoxical association between higher BMI and improved clinical outcomes in cardiovascular diseases, including heart failure and CS, is more prominent in males than females, especially in the Asian population.15),16),17),18) The underlying mechanism is unclear; however, Meyer et al.19) reported that in patients with heart failure, the levels of neurohormones related to inflammation and extracellular matrix remodeling were significantly lower in female patients than male patients. Our findings correspond well with those of earlier studies that established the clinical implications of the obesity paradox in male patients suffering from diverse cardiovascular diseases. By analyzing long-term clinical outcomes in the setting of CS stratified by etiology, our study provides valuable information for predicting the prognosis of CS patients.
In our study, CAG was less frequently performed in females. Previous studies have also reported that females received significantly less CAG than their male counterparts among CS patients.3),20) Subsequently, the lower use of CAG in females was followed by less frequent revascularization and higher in-hospital mortality.20) Reassuringly, in our registry, revascularization rates were equitable, and females had similar risks of in-hospital mortality compared to males, highlighting the importance of interventions actually performed in female CS patients. The lower use of CAG in female patients is often attributed to the atypical presentation of myocardial ischemia symptoms, such as atypical chest pain, dyspnea, or general weakness, which can lead to misdiagnosis or delayed revascularization.21) Considering the hemodynamically diverse status of CS patients, with a majority of clinical events occurring early in diagnosis, the atypical presentation in females may result in adverse clinical outcomes.21) Efforts toward providing more intensive care for females could improve clinical outcomes in CS patients. While we found equitable in-hospital management with no observed sex differences in clinical outcomes, our findings underscore the need for tailored management strategies for CS patients during both in-hospital and post-discharge periods, considering sex differences and the impact of obesity through detailed analyses. The obesity paradox in males, along with the lower utilization of CAG, renal replacement therapy, mechanical ventilation, and MCS in females, could be considerable factors influencing clinical outcomes. Future research should focus on understanding the mechanisms underlying these observations and developing interventions that address these issues. By doing so, all CS patients can receive optimal care, leading to better long-term prognoses.
Despite the strengths of this study as a large, multicenter investigation using a recent real-world CS registry with minimal exclusion criteria, several limitations should be considered. First, as a non-randomized, retrospective, observational study, unmeasured confounding factors and selection bias may have influenced the results. While we employed multivariable Cox regression and propensity-score matching to minimize these biases, adjustments for unmeasured variables were not possible. Second, our registry did not capture detailed data on newly diagnosed disease or medical treatment, including the initiation, type, and dosage of heart failure medications during hospitalization or after discharge. Consequently, we were unable to assess the impact of guideline-directed medical therapy on clinical outcomes, as emphasized in prior trials like PIONEER-HF.22) Additionally, echocardiographic and hemodynamic parameters at enrollment were not systematically collected, limiting our ability to evaluate baseline cardiac function. Third, the non-ICMP group had a relatively small sample size, which may have reduced statistical power and limited the generalizability of findings in this subgroup. Larger cohort in the future studies will be necessary to further investigate sex-based differences in this population. Fourth, follow-up data on serial NT-proBNP levels or left ventricular ejection fraction measurements at 6–12 months post-shock were unavailable, preventing us from assessing the degree of myocardial recovery. Furthermore, the 12-month follow-up duration may have been insufficient to fully evaluate long-term sex differences in prognosis. Finally, our study was constrained by limited statistical power, particularly within the prospective cohort. Future multicenter studies with larger, international populations—such as the RESCUE II registry—will be essential for validating these findings.
In this large-scale, multicenter study of patients with CS, we found no significant sex differences in in-hospital management or in clinical outcomes, both in-hospital and at 12 months, regardless of the etiology of the shock. These findings suggest that sex may not be a significant factor in determining the long-term prognosis of patients with CS. This study provides important insights into the management and outcomes of CS, emphasizing the need for continued efforts to ensure equitable treatment for all patients. Future research should focus on potential therapeutic interventions to improve outcomes across all patient subgroups.
Footnotes
Funding: The authors received no financial support for the research, authorship, and/or publication of this article.
Conflict of Interest: The authors have no financial conflicts of interest.
Data Sharing Statement: The data generated in this study is available from the corresponding author upon reasonable request.
- Conceptualization: Park IH, Kim CH, Oh JH, Lee WS, Yang JH, Gwon HC, Jang WJ.
- Data curation: Park IH, Kim CH, Oh JH, Lee WS, Yang JH, Gwon HC, Jang WJ.
- Formal analysis: Park IH, Kim CH, Jang WJ.
- Investigation: Park IH, Kim CH, Lee WS, Jang WJ.
- Methodology: Park IH, Kim CH, Yang JH, Gwon HC.
- Project administration: Park IH, Kim CH, Oh JH, Yang JH, Gwon HC, Jang WJ.
- Resources: Oh JH, Gwon HC.
- Supervision: Oh JH, Lee WS, Yang JH, Gwon HC, Jang WJ.
- Validation: Oh JH, Lee WS, Yang JH, Jang WJ.
- Visualization: Park IH, Kim CH.
- Writing - original draft: Park IH, Kim CH, Jang WJ.
- Writing - review & editing: Park IH, Kim CH, Lee WS, Yang JH, Gwon HC, Jang WJ.
SUPPLEMENTARY MATERIALS
Percent standardized differences of variables among ICMP group of the unadjusted and propensity-score matched cohort, and baseline clinical characteristics and in-hospital managements in the ICMP group of the propensity-score matched cohort
Percent standardized differences of variables among non-ICMP group of the unadjusted and propensity-score matched cohort, and baseline clinical characteristics and in-hospital managements in the non-ICMP group of the propensity-score matched cohort
In-hospital clinical outcomes
Thirty-day in-hospital clinical outcomes
Twelve-month follow-up clinical outcomes
Twelve-month follow-up clinical outcomes for patients surviving to discharge
Baseline clinical characteristics and in-hospital managements of patients surviving to discharge
Schematic illustration of study cohort selection.
Time-to-event Kaplan-Meier curves for 12-month cumulative incidence of (A) recurrent myocardial infarction, (B) unplanned repeat revascularization, (C) readmission due to heart failure, and (D) cerebrovascular accidents in the ischemic cardiomyopathy population.
Time-to-event Kaplan-Meier curves for 12-month cumulative incidence of (A) readmission due to heart failure and (B) cerebrovascular accidents in the non-ischemic cardiomyopathy population.
References
- 1.van Diepen S, Katz JN, Albert NM, et al. Contemporary management of cardiogenic shock: a scientific statement from the American Heart Association. Circulation. 2017;136:e232–e268. doi: 10.1161/CIR.0000000000000525. [DOI] [PubMed] [Google Scholar]
- 2.Osman M, Syed M, Kheiri B, et al. Age stratified sex-related differences in incidence, management, and outcomes of cardiogenic shock. Catheter Cardiovasc Interv. 2022;99:1984–1995. doi: 10.1002/ccd.30177. [DOI] [PubMed] [Google Scholar]
- 3.Elgendy IY, Wegermann ZK, Li S, et al. Sex differences in management and outcomes of acute myocardial infarction patients presenting with cardiogenic shock. JACC Cardiovasc Interv. 2022;15:642–652. doi: 10.1016/j.jcin.2021.12.033. [DOI] [PubMed] [Google Scholar]
- 4.Lozano-Jiménez S, Iranzo-Valero R, Segovia-Cubero J, et al. Gender differences in cardiogenic shock patients: clinical features, risk prediction, and outcomes in a hub center. Front Cardiovasc Med. 2022;9:912802. doi: 10.3389/fcvm.2022.912802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Yang JH, Choi KH, Ko YG, et al. Clinical characteristics and predictors of in-hospital mortality in patients with cardiogenic shock: results from the RESCUE registry. Circ Heart Fail. 2021;14:e008141. doi: 10.1161/CIRCHEARTFAILURE.120.008141. [DOI] [PubMed] [Google Scholar]
- 6.Yamazaki Y, Oba K, Matsui Y, Morimoto Y. Vasoactive-inotropic score as a predictor of morbidity and mortality in adults after cardiac surgery with cardiopulmonary bypass. J Anesth. 2018;32:167–173. doi: 10.1007/s00540-018-2447-2. [DOI] [PubMed] [Google Scholar]
- 7.Cutlip DE, Windecker S, Mehran R, et al. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation. 2007;115:2344–2351. doi: 10.1161/CIRCULATIONAHA.106.685313. [DOI] [PubMed] [Google Scholar]
- 8.Kunadian V, Qiu W, Bawamia B, Veerasamy M, Jamieson S, Zaman A. Gender comparisons in cardiogenic shock during ST elevation myocardial infarction treated by primary percutaneous coronary intervention. Am J Cardiol. 2013;112:636–641. doi: 10.1016/j.amjcard.2013.04.038. [DOI] [PubMed] [Google Scholar]
- 9.Cenko E, van der Schaar M, Yoon J, et al. Sex-related differences in heart failure after ST-segment elevation myocardial infarction. J Am Coll Cardiol. 2019;74:2379–2389. doi: 10.1016/j.jacc.2019.08.1047. [DOI] [PubMed] [Google Scholar]
- 10.Collado-Lledó E, Llaó I, Rivas-Lasarte M, et al. Clinical picture, management and risk stratification in patients with cardiogenic shock: does gender matter? BMC Cardiovasc Disord. 2020;20:189. doi: 10.1186/s12872-020-01467-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Yan I, Schrage B, Weimann J, et al. Sex differences in patients with cardiogenic shock. ESC Heart Fail. 2021;8:1775–1783. doi: 10.1002/ehf2.13303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mansur AP, Pereira-Barretto AC, Del Carlo CH, et al. Sex differences in prognosis of heart failure due to ischemic and nonischemic cardiomyopathy. J Clin Med. 2023;12:5323. doi: 10.3390/jcm12165323. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lainščak M, Milinković I, Polovina M, et al. Sex- and age-related differences in the management and outcomes of chronic heart failure: an analysis of patients from the ESC HFA EORP Heart Failure Long-Term Registry. Eur J Heart Fail. 2020;22:92–102. doi: 10.1002/ejhf.1645. [DOI] [PubMed] [Google Scholar]
- 14.Iorga A, Cunningham CM, Moazeni S, Ruffenach G, Umar S, Eghbali M. The protective role of estrogen and estrogen receptors in cardiovascular disease and the controversial use of estrogen therapy. Biol Sex Differ. 2017;8:33. doi: 10.1186/s13293-017-0152-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Komukai K, Minai K, Arase S, et al. Impact of body mass index on clinical outcome in patients hospitalized with congestive heart failure. Circ J. 2012;76:145–151. doi: 10.1253/circj.cj-11-0727. [DOI] [PubMed] [Google Scholar]
- 16.Hong S, Lee JH, Kim KM, et al. Is there a sex-related difference in the obesity paradox in systolic heart failure? Sex-related difference in the obesity paradox. Yonsei Med J. 2018;59:57–62. doi: 10.3349/ymj.2018.59.1.57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lee SY, Kim HL, Kim MA, et al. Obesity paradox in Korean male and female patients with heart failure: a report from the Korean Heart Failure Registry. Int J Cardiol. 2021;325:82–88. doi: 10.1016/j.ijcard.2020.10.013. [DOI] [PubMed] [Google Scholar]
- 18.Kwon W, Lee SH, Yang JH, et al. Impact of the obesity paradox between sexes on in-hospital mortality in cardiogenic shock: a retrospective cohort study. J Am Heart Assoc. 2022;11:e024143. doi: 10.1161/JAHA.121.024143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Meyer S, van der Meer P, van Deursen VM, et al. Neurohormonal and clinical sex differences in heart failure. Eur Heart J. 2013;34:2538–2547. doi: 10.1093/eurheartj/eht152. [DOI] [PubMed] [Google Scholar]
- 20.Vallabhajosyula S, Ya’Qoub L, Singh M, et al. Sex disparities in the management and outcomes of cardiogenic shock complicating acute myocardial infarction in the young. Circ Heart Fail. 2020;13:e007154. doi: 10.1161/CIRCHEARTFAILURE.120.007154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Mehta LS, Beckie TM, DeVon HA, et al. Acute myocardial infarction in women: a scientific statement from the American Heart Association. Circulation. 2016;133:916–947. doi: 10.1161/CIR.0000000000000351. [DOI] [PubMed] [Google Scholar]
- 22.Velazquez EJ, Morrow DA, DeVore AD, et al. Angiotensin-neprilysin inhibition in acute decompensated heart failure. N Engl J Med. 2019;380:539–548. doi: 10.1056/NEJMoa1812851. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Percent standardized differences of variables among ICMP group of the unadjusted and propensity-score matched cohort, and baseline clinical characteristics and in-hospital managements in the ICMP group of the propensity-score matched cohort
Percent standardized differences of variables among non-ICMP group of the unadjusted and propensity-score matched cohort, and baseline clinical characteristics and in-hospital managements in the non-ICMP group of the propensity-score matched cohort
In-hospital clinical outcomes
Thirty-day in-hospital clinical outcomes
Twelve-month follow-up clinical outcomes
Twelve-month follow-up clinical outcomes for patients surviving to discharge
Baseline clinical characteristics and in-hospital managements of patients surviving to discharge
Schematic illustration of study cohort selection.
Time-to-event Kaplan-Meier curves for 12-month cumulative incidence of (A) recurrent myocardial infarction, (B) unplanned repeat revascularization, (C) readmission due to heart failure, and (D) cerebrovascular accidents in the ischemic cardiomyopathy population.
Time-to-event Kaplan-Meier curves for 12-month cumulative incidence of (A) readmission due to heart failure and (B) cerebrovascular accidents in the non-ischemic cardiomyopathy population.