Rethinking early risk stratification in heart failure-related cardiogenic shock

Cardiogenic shock (CS) remains among the most devastating complications in cardiovascular medicine, with mortality rates exceeding 40 % despite advances in intensive care and circulatory support [1]. While the majority of research has focused on myocardial infarction–related shock, heart failure–related cardiogenic shock (HF-CS) is increasingly recognized as a distinct and highly prevalent entity with its own trajectory, management challenges, and prognostic uncertainty. In this issue of IJC Heart & Vasculature, Ris and colleagues provide an important contribution by interrogating the predictive performance of the Cardiogenic Shock Working Group–modified Society for Cardiovascular Angiography and Interventions (CSWG-SCAI) staging system in early-stage HF-CS. Their results highlight both the promise and the shortcomings of this widely adopted classification [2]. The investigators retrospectively analyzed 208 patients with HF-CS across two Dutch tertiary centers, with a median age of 73 years and a predominance of acute-on-chronic decompensation. Importantly, patients with acute myocardial infarction, out-of-hospital cardiac arrest (OHCA), or invasive mechanical ventilation at baseline were excluded, thereby defining an “early-stage” cohort that differs from the mixed, critically ill populations in which SCAI staging has previously been validated [3,4]. At the time of HF-CS diagnosis, nearly 70 % of patients were classified as stage A or B, and strikingly, SCAI stage at this initial time point bore no relation to in-hospital mortality. In contrast, reassessment at 24 and 48 h, as well as consideration of the maximum stage achieved during hospitalization, demonstrated robust associations with outcome. Deterioration in stage within the first 24 h emerged as a particularly powerful marker of poor prognosis. The authors further tested whether the addition of urine output—a simple and widely monitored parameter—could enhance risk prediction. Although statistically significant, the incremental gain was modest at best, with the area under the ROC curve improving from 0.67 to 0.70 at 48 h [2]. These observations are both reassuring and unsettling. On the one hand, the findings confirm that the CSWG-SCAI framework retains prognostic relevance when applied dynamically and serially, particularly after the first day of shock management [5]. On the other, the inability of the classification to discriminate outcomes at the very early phase of HF-CS diagnosis—the stage at which clinical decision-making is most urgent—raises fundamental concerns about its utility as an early risk stratification tool. Indeed, patients classified as stage A paradoxically exhibited high mortality, challenging both the staging system and the clinical adjudication that led to their inclusion. Several explanations are plausible: incomplete hemodynamic profiling leading to misclassification, premature initiation of inotropes in patients not yet in overt shock, or, conversely, that the current criteria simply fail to capture the true pathophysiology of early HF-CS, where subtle organ hypoperfusion precedes overt hypotension or lactate elevation [6]. Retrospective assignment of SCAI stages based on chart data may also have compounded misclassification, as invasive hemodynamic measurements were not systematically obtained [3,4]. Yet such limitations reflect the clinical reality in which these staging systems are meant to operate: decisions are often made with incomplete data, and tools that lack robustness under these circumstances may have limited bedside value. Still, the 42 % mortality observed in this supposedly less critically ill cohort emphasizes the gravity of HF-CS and the inadequacy of our current stratification methods [2,3,7]. The paradox of high mortality even among patients staged as A or B should serve as a cautionary signal: reliance on CSWG-SCAI in isolation may obscure the true risk in early disease. What is the conclusion from this analysis? First, initial staging at diagnosis may be misleading, while changes within the first 24 h—the so-called “golden day of shock”—carry profound prognostic implications [7]. This aligns with recent work demonstrating that the trajectory of shock, rather than the static stage at a single time point, dictates outcomes [3,5,7]. Second, the limited additive value of urine output, despite its intuitive appeal, suggests that simply expanding existing criteria with single parameters may be insufficient. More comprehensive multidimensional models integrating renal function, biomarkers such as NT-proBNP or troponin, echocardiographic indices, and invasive hemodynamic data may ultimately be required [8]. Third, the findings highlight the importance of external validation across different health systems and patient subsets. The ongoing ALTSHOCK-2 trial, which recently questioned the benefit of intra-aortic balloon support in HF-CS, illustrates how treatment strategies differ markedly from AMI-CS and require tailored risk tools [9]. From a methodological perspective, it is important to note that the use of maximum SCAI stage, while strongly associated with mortality, is biased. Patients who survive longer naturally accumulate worse parameters, inflating the apparent predictive power. As the authors correctly caution, maximum stage is of limited clinical utility [2]. Instead, the emphasis should be on early, prospective markers that can guide real-time decision-making. Whether advanced monitoring, machine learning models, or novel biomarkers will fulfill this need remains to be tested. For the practicing cardiologist or intensivist, the study suggests that instead of a snap-shot at presentation a close re-evaluation at 24 h is essential, and deterioration at this juncture should prompt urgent reassessment of management, including consideration of advanced therapies, transfer to specialized centers, or initiation of mechanical support. Ultimately, CSWG-SCAI provides a standardized language for communication, but it cannot substitute for nuanced clinical judgment and comprehensive assessment. Prospective, multicenter validation of CSWG-SCAI in well-defined HF-CS cohorts is urgently needed, with standardized data collection including invasive hemodynamics and renal function. Integration of biomarkers such as NT-proBNP, which in the present cohort showed predictive value but was excluded due to timing inconsistencies, could strengthen risk models [3,7,8]. Efforts to refine the definition of “early-stage” HF-CS are also essential; reliance on exclusion criteria such as absence of cardiac arrest or mechanical ventilation may be insufficient. Finally, as new trials continue to challenge the efficacy of mechanical support in HF-CS, improved stratification tools will be critical to identify which patients, if any, may benefit from such interventions [9,10].

Article type

Invited Editorial to “Predictive value of cardiogenic shock working group-modified SCAI criteria in early-stage heart failure-related cardiogenic shock” (IJCHA-D-25-00563R1).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.