Cardiac dysfunction in septic patients admitted to the intensive care unit (ICU)—referred to as sepsis-induced myocardial dysfunction or septic cardiomyopathy (SCM)—is usually classified as left ventricular (LV), right ventricular (RV), or biventricular systolic dysfunction (LVSD/RVSD). In addition to conventional LV functional assessment using ejection fraction, myocardial strain measured by speckle-tracking echocardiography has emerged as a robust marker of intrinsic cardiac function, detecting systolic and diastolic dysfunction up to 50% more frequently.1

The presence of harmful variants in genes associated with cardiomyopathies has been reported in forms of ventricular dysfunction or dilated cardiomyopathy (DCM) that were previously considered predominantly acquired or nongenetic. This is particularly evident for variants in titin (TTN) in the context of peripartum cardiomyopathy, cancer therapy–related cardiotoxicity, alcohol-associated cardiomyopathy,2 and myocarditis.3

Therefore, we hypothesized that a similar genetic predisposition may exist in SCM and investigated whether rare genetic variants in cardiomyopathy-associated genes could be detected in this context. Such findings may have implications for both clinical and familial management.

A single-center prospective observational study was conducted in a general ICU between October 2022 and December 2023. The study was approved by the ethics committees of Hospital Garcia de Orta and the Instituto de Medicina Molecular (T01/2022), and written informed consent was obtained from all participants. The study was registered on ClinicalTrials.gov (NCT05552521) on 15 September 2022.

Patients aged ≥18 years who met septic shock criteria according to the Surviving Sepsis Campaign and who had cardiac dysfunction were consecutively recruited. A comprehensive echocardiogram was performed within 24hours of ICU admission and analyzed offline using dedicated software (GE EchoPAC PC v206, GE Healthcare, United States). Cardiac dysfunction was classified based on current guidelines. LV systolic dysfunction was defined as an LV ejection fraction (LVEF)<50% or a global longitudinal strain (GLS) >–15%. Exclusion criteria included pregnancy, congenital heart disease, prior LV or RV dysfunction, valve prostheses, severe valvular disease, atrial fibrillation, acute coronary syndrome, acute pulmonary embolism, and inadequate image quality. All patients had previous echocardiograms from routine ambulatory health check-ups within the year prior to admission, none of which showing an LVEF <50%.

Whole-exome sequencing was performed at the Institute of Molecular Medicine, Lisbon. Analysis focused on a panel of 281 genes associated with cardiomyopathies, including genes linked to metabolic conditions and all genes with definitive, strong, or moderate evidence for association with DCM, as classified by ClinGen. These included: definitive: BAG3, DES, FLNC, LMNA, MYH7, PLN, RBM20, SCN5A, TNNC1, TNNT2, TTN; strong: DSP; moderate: ACTC1, ACTN2, JPH2, NEXN, TNNI3, TPM1, VCL. Variants were filtered for a minor allele frequency <0.0001 and classified according to the American College of Medical Genetics and Genomics (ACMG) criteria.4 Copy-number variation was also analysed. Additionally, we analyzed a group of genes recently suggested to be associated with septic cardiomyopathy: ADGRG1, BCL2A1, BCL6, CD44, FTH1, GAB2, IL1R1, IL6ST, ING3, JAK3, KLRG1, MMP9, PRF1, SERPINE1, TGIF1, and TPT1.5

All demographic, clinical, hemodynamic, respiratory, fluid balance, vasopressor use, and laboratory data were prospectively collected, along with 30-day mortality.

Continuous variables are presented as medians with interquartile ranges and compared between 2 groups using the Mann-Whitney U test, as appropriate. Categorical variables are expressed as proportions and compared using the Fisher exact test.

Of the initial 151 patients assessed, 71 met exclusion criteria, 4 had insufficient DNA quality for genetic analysis, and 6 had isolated diastolic dysfunction. The remaining 70 patients had systolic dysfunction, as defined by either LVEF or LV-GLS criteria; among them, 45 patients (64%) had an LVEF<50%.

The baseline clinical and echocardiographic characteristics of the 70 patients with LVSD, obtained within the first 24hours of ICU admission, are presented in table 1 and table 2, respectively.

The median age was 66 years (IQR: 52-74), with 41% of patients being female. The median Acute Physiology and Chronic Health Evaluation II (APACHE II) score was 20 (IQR: 13-27), and the median Simplified Acute Physiology Score II (SAPS II) was 46 (IQR: 31-61). Invasive mechanical ventilation was required in 62% of patients, 41% received renal replacement therapy, and the 30-day mortality rate was 42%.

Patients with LVEF <50% had higher SAPS II scores, a greater need for renal replacement therapy, and elevated levels of high-sensitivity troponin and N-terminal pro B-type natriuretic peptide (NT-proBNP). They also had a lower cardiac index and reduced central venous oxygen saturation. Echocardiographically, these patients showed a larger LV end-systolic diameter (37 [33-42] vs 27 [24-42] mm; P=.017) and LV end-systolic volume (58 [50-63] vs 38 [31-48] mL; P<.001), lower LV-GLS (−11 [−9 to −13] vs -14 [−12 to −19] %; P<.001) and worse right ventricle systolic function.

No pathogenic or likely pathogenic variants were identified in the analyzed DCM-associated genes, genes related to inherited metabolic conditions associated with cardiomyopathy, or genes recently proposed as candidates for an association with septic cardiomyopathy.

This is the first study to investigate the contribution of inherited cardiomyopathy-related genetic variation in a septic shock population with cardiac dysfunction. The absence of pathogenic or likely pathogenic variants in 70 patients with systolic dysfunction—including 45 with LVEF <50%—suggests a potential lack of this specific genetic predisposition in SCM. However, given the limited sample size, validation in larger, multicenter cohorts is necessary to confirm these findings.

In contrast, a genetic predisposition has been demonstrated in other forms of non-ischemic cardiomyopathy previously considered acquired. In alcoholic cardiomyopathy, heterozygous rare protein-altering variants in 9 genes associated with inherited DCM were identified in 19 of 141 patients (13.5%), a frequency significantly higher than in controls and comparable to that seen in genetic DCM. Similarly, titin truncating variants (TTNtv) were identified in 172 cases of peripartum cardiomyopathy, indicating a shared genetic predisposition with familial DCM.2 Rare TTNtv have also been associated with an increased risk of cancer therapy-induced cardiomyopathy.2 DCM- or arrhythmogenic cardiomyopathy-associated variants were identified in 8% of 336 patients with acute myocarditis,3 with TTNtv predominance in those with reduced left ventricular ejection fraction.

An association between sepsis, SCM, and immune response genes has been previously reported.5 Although SCM can be profound, it is usually reversible in survivors.6 Our findings might therefore reflect similarities with other clinical scenarios of stress-induced LV dysfunction (eg, tako-tsubo cardiomyopathy) in which the recovery of LV function and normalization of LV size can occur.6 In these entities, there is likewise no established contribution from cardiomyopathy-associated genetic variants. In summary, our results suggest that cardiomyopathy-related genetic variation does not play a significant role in the pathogenesis of SCM.