Cardiogenic shock (CS), characterized by critically reduced cardiac output leading to systemic circulatory failure, multi-organ hypoperfusion, and hypoxia, carries a mortality rate exceeding 40% despite therapeutic advances (1). As a complication of 3%–13% of acute myocardial infarction (AMI) cases, acute myocardial infarction-related cardiogenic shock (AMI-CS) remains one of the most predominant etiologies associated with high mortality (2–4). Current guidelines emphasize coronary revascularization, vasopressors, and mechanical circulatory support (MCS) to restore tissue oxygenation in AMI-CS patients (5). However, only immediate culprit vessel revascularization and Impella device deployment have demonstrated consistent survival benefits (2, 6, 7), with mortality persisting to be substantial (8, 9).

Hemoglobin (Hb), the principal oxygen-carrying component of erythrocytes, plays a pivotal role in maintaining systemic oxygen delivery–demand equilibrium (10). While anemia has been independently associated with worsened outcomes in both general CS populations and AMI patients without shock (11–14), its prognostic significance in AMI-CS remains underexplored. Existing studies are limited by reliance on single-point Hb measurements (15), neglecting the dynamic nature of Hb fluctuations during hospitalization and their cumulative impact on clinical outcomes.

To address this gap, our study investigated the association between time-weighted average hemoglobin (TWA-Hb)—a novel metric quantifying cumulative Hb exposure—and all-cause mortality in AMI-CS patients. This approach enables comprehensive evaluation of Hb trajectory's prognostic value beyond conventional static assessments.

After screening the data of 1,085 patients with AMI-CS, a total of 765 adult individuals were finally recruited in the present study, with a median age of 72.01 years (IQR: 71.11–72.90) and male patients accounting for 60.9% (n = 466). Of the study cohort, the 28-day, 90-day, 6-month, and 1-year mortality were recorded at 44.6%, 47.5%, 49.9%, and 53.2% of the populations, respectively, and the Hb level of the overall cohort was 10.33 g/dl (IQR: 10.21–10.46). The flow diagram of patient selection is shown in Figure 1.

Our study demonstrates that lower TWA-Hb levels are significantly associated with elevated long-term mortality in patients with AMI-CS, particularly among elderly individuals and those without DM. To our knowledge, this is the first investigation to establish TWA-Hb—a dynamic metric reflecting cumulative Hb exposure—as an independent prognostic marker in AMI-CS populations.

Existing studies concerning the correlation of Hb with clinical outcomes almost utilized Hb at a single time point and ignored the time-varying characteristic of Hb along with the process of diagnosis and treatment. Notably, anemia is prevalent among AMI-CS patients (16, 17), which might either present on admission or develop during hospitalization as a result of complicated treatments. Moreover, anemia could often trigger RBC transfusions (18). As such, defining Hb status only by using its once-off value could cause potential regression dilution bias and ultimately affect the precision of findings. Lately, an increasing number of studies have investigated the association of cumulative exposure to biological measures (e.g., blood pressure and cholesterol) with the prognosis of cardiovascular disease (19, 20). However, the cumulative effect of Hb on outcomes has not been well studied. A previous study has described that in-hospital Hb drop was apparently correlated with an increased risk for long-term mortality in AMI patients (18). Furthermore, Kalra et al. (21) conducted a multicenter cohort enrolling patients from 45 countries and found that patients with anemia on admission appeared to have no increased mortality risk if their anemia normalized over time. Overall, even during a very short stay in the hospital, the condition evolution of AMI-CS varied rapidly and diversely (1, 22), so the comprehensive cumulative exposure to Hb should be considered when assessing AMI-CS patients' outcomes.

Using the TWA-Hb in the present analysis, we found that lower TWA-Hb presented a strong impact on long-term mortality among patients with AMI-CS, presenting increased mortality at 90-day, 6-month, and 1-year follow-ups. To this day, little is known about the impact of Hb concentrations or anemia presence on the long-term prognosis in patients suffering from AMI-CS. In contrast, concordant with our findings, a study in previous years has linked the improvement of anemia to prolonged survival for non-shock AMI populations (23). Another study also explored that a Hb drop of >3 g/dl during hospitalization was correlated with increased 1-year and 5-year mortality of AMIs (18, 24). Nonetheless, the specific mechanism of lower Hb exposure worsening long-term survival remains inconclusive. One may speculate that inadequate oxygen carriers could impair cellular oxidative metabolism and exaggerate myocardial ischemia, thus promoting the extension of infarcted size (25, 26), as confirmed using SPECT imaging by Dutsch et al. (18) that larger myocardial infarctions were found in those with a Hb decrease of ≥3 g/dl during hospitalization. In the setting of CS, anemia-mediated reduced myocardium oxygen supply would be further jeopardized by CS-associated hypoperfusion, which was more likely to cause a lack of recovery of cardiac pump function after the acute state. In line, a retrospective study including 39,922 STEMI individuals exhibited that baseline Hb <14 g/dl was associated with a higher risk of future heart failure (27). Given that low LVEF has been recognized as one of the pivotal elements related to poor long-term prognosis of cardiovascular diseases (28, 29), we deemed that the reduced myocardial salvage and deteriorated heart function might primarily account for the increased long-term mortality in relation to low levels of TWA-Hb in populations with AMI-CS.

We also evaluated the association between TWA-Hb and short-term mortality (28 days), obtaining insignificant results consistent with those in previous studies regarding STEMI populations but inconsistent with those concerning CS patients. A retrospective study enrolling 1,919 STEMI patients undergoing PCI reported that anemic status did not affect the in-hospital mortality of these patients (30). More recently, another study involving 3,071 STEMI individuals also confirmed that anemia upon admission was not correlated with the risk of in-hospital mortality (31). However, among anemic CS (from any cause) patients, Obradovic et al. (15) found a significantly higher HR for 30-day mortality compared with those without anemia. During this phase, the Hb levels may play a less dominant role in determining immediate survival, as the primary focus is on myocardial reperfusion and circulatory stabilization (32). However, as the patient moves into the sub-acute and chronic phases of recovery, the importance of TWA-Hb and its influence on oxygen delivery to tissues becomes more pronounced. Anemia, even if not immediately life-threatening, can impair oxygen delivery to the myocardium and other organs, potentially impeding recovery and leading to a higher risk of mortality over longer time frames. The chronic reduction in oxygen-carrying capacity can exacerbate myocardial stunning, impair neovascularization, and lead to a vicious cycle of worsening cardiac function and systemic complications. Therefore, while the immediate survival after AMI-CS heavily benefits from timely revascularization and optimal MCS, the longer-term outcomes might be more closely tied to adequate oxygen delivery. Our findings further suggest that TWA-Hb may be a more significant predictor of long-term survival, highlighting the importance of managing anemia in the post-AMI-CS period to improve patient outcomes.

There are still some limitations. First, due to its post hoc characteristic, we could not conclude causality between TWA-Hb and all-cause mortality, so further prospective cohorts are required. Second, we could not completely rule out the presence of residual bias, although we performed multi-adjustments for potential confounders. For instance, AMI patients with more severe initial shock were associated with a higher risk of mortality (33). Even after maximal adjustment for confounding factors such as LAC, pH, MAP, the use of vasopressors, and IABP as surrogates for assessing shock severity, the absence of certain data (such as cardiac index and pulmonary capillary wedge pressure) necessary for SCAI classification may have resulted in less accurate risk stratification. Moreover, MCS, such as VA-ECMO and Impella, might hold the potential benefits in improving survival in patients with AMI-CS (7), allowing for myocardial recovery or bridging to decision-making, including the possibility of advanced heart failure therapies. The adoption of these advanced devices might have contributed to better outcomes in patients with AMI-CS, which could have been underrepresented in our initial analysis. Baseline iron status could influence Hb levels and affect the organism's capacity to manage oxidative stress and inflammation, both of which are known to impact prognoses (34). Notably, a history of transfusions might correlate with more severe anemia and other health complications, potentially elevating mortality. Thus, there is a clear need for prospective cohorts with more detailed records. Third, patients enrolled come from a single center, so multicenter studies are necessary to verify the generalizability of these findings. Fourth, data presented in this study, which covers the period from 2008 to 2019, might not reflect the more recent changes in PCI strategies and the broader adoption of advanced MCS devices such as VA-ECMO and Impella. The evolution of PCI techniques, including the use of more sophisticated stenting materials, improved antiplatelet therapies, and enhanced imaging guidance, could potentially improve procedural success rates and reduce complications. Similarly, the increased utilization of MCS devices such as VA-ECMO and Impella might have altered the landscape of AMI-CS treatment, offering new possibilities for patient recovery that are not minutely captured in our study. Therefore, while our study provides valuable historical context, ongoing and future studies with more current data are essential to ensure that clinical guidelines and patient care strategies remain up to date and reflect the latest evidence-based practices. Finally, the function of TWA may not adequately differentiate between patients who experienced a drop in Hb from a high baseline vs. those who had an increase in Hb from a low baseline, preventing a more nuanced analysis of Hb changes in the context of bleeding complications.

TWA-Hb is an indicator of long-term all-cause mortality among patients suffering from AMI-CS. Serial assessment of TWA-Hb in AMI-CS routine clinical management may improve risk stratification and could thereby guide clinical decision-making. However, additional prospective investigations are still warranted to validate our results.