Section Structure
1. Highlights
2. Clinical Background
3. Study Design and Methods
4. Key Results
5. Clinical Interpretation
6. Strengths and Limitations
7. Implications for Practice and Research
8. Funding and Trial Registration
9. References
Highlights
Among patients with acute myocardial infarction (AMI) and anemia enrolled in the MINT trial, critically ill patients had substantially higher 30-day rates of death and death/myocardial infarction (MI) than noncritically ill patients.
When patients were stratified by ICU status at randomization, a restrictive transfusion strategy was associated with numerically worse 30-day outcomes than a liberal strategy in both groups, but the differences were not statistically significant.
Interaction testing did not support a meaningful difference in treatment effect between critically ill and noncritically ill patients, suggesting that the overall MINT signal favoring a liberal strategy applies across illness severity categories.
These findings are clinically important because transfusion thresholds in AMI remain a major area where generalized restrictive transfusion policies may not fully fit a high-risk ischemic population.
Clinical Background
Anemia is common in hospitalized adults and especially frequent in intensive care units. In most critically ill populations, randomized trials have supported a restrictive transfusion strategy, usually using a hemoglobin threshold around 7 g/dL, based on evidence that it is at least as safe as more liberal transfusion and may reduce exposure to blood products. This evidence base has strongly influenced hospital transfusion practice.
Patients with AMI represent a biologically and clinically distinct subgroup. In myocardial ischemia, reduced oxygen-carrying capacity may plausibly worsen the imbalance between myocardial oxygen supply and demand. That concern has long complicated transfusion decision-making in AMI, particularly when hemoglobin is low but not profoundly low. Older data, including subgroup observations from the Transfusion Requirements in Critical Care (TRICC) trial, raised the possibility that patients with ischemic heart disease might not benefit from the same restrictive approach used in general critical care.
The parent MINT trial directly addressed this uncertainty by randomizing patients with AMI and anemia to restrictive or liberal red cell transfusion strategies. The present report is a prespecified or protocol-driven subgroup analysis examining whether the relative effects of these strategies differ in critically ill versus noncritically ill patients. This question is clinically relevant because ICU status often influences transfusion habits, perceived risk, and tolerance for anemia.
Study Design and Methods
This study was a subgroup analysis of the Myocardial Ischemia and Transfusion (MINT) trial, a parallel, two-arm randomized controlled trial conducted at 144 sites in the United States, Canada, France, Brazil, New Zealand, and Australia. Adult patients with AMI and anemia were eligible. For this analysis, patients were classified as critically ill if they were being treated in an ICU at the time of randomization; all others were classified as noncritically ill.
Participants were randomly assigned to one of two transfusion strategies. In the restrictive strategy, transfusion was triggered when hemoglobin fell below 7 to 8 g/dL. In the liberal strategy, transfusion was given when hemoglobin fell below 10 g/dL. As in the parent trial, the comparison reflects a strategy-based approach rather than a single-unit transfusion decision.
The principal outcomes assessed in this subgroup analysis were 30-day all-cause death and the composite of death or recurrent MI. The investigators compared event rates within the critically ill and noncritically ill strata and tested for interaction to determine whether ICU status modified the relative effect of transfusion strategy.
Of 3504 patients enrolled in MINT, 1679 patients, or 47.9%, were critically ill at randomization. This nearly even split provides a clinically useful window into whether severity of illness alters the balance of transfusion risk and benefit in AMI.
Key Results
Baseline risk by critical illness status
Critical illness was associated with substantially worse 30-day outcomes regardless of transfusion assignment. Death occurred in 11.9% of critically ill patients versus 6.5% of noncritically ill patients. The composite of death or MI occurred in 18.9% versus 12.7%, respectively. These differences underscore that ICU status identified a much higher-risk AMI population, as expected.
Effect of restrictive versus liberal transfusion in noncritically ill patients
Among noncritically ill patients, the restrictive strategy was associated with a nonsignificant 9% relative increase in death compared with the liberal strategy, with a risk ratio (RR) of 1.09 and a 95% confidence interval (CI) of 0.77 to 1.54. For the composite of death or MI, the restrictive strategy was likewise associated with a nonsignificant 9% relative increase, RR 1.09, 95% CI 0.85 to 1.38.
These estimates do not establish superiority of the liberal strategy with statistical certainty in this subgroup. However, the direction of effect is consistent with the overall MINT finding that a liberal transfusion approach may be preferable in AMI with anemia.
Effect of restrictive versus liberal transfusion in critically ill patients
Among critically ill patients, the signal against restrictive transfusion was numerically stronger. The restrictive strategy was associated with a 24% relative increase in death, RR 1.24, 95% CI 0.95 to 1.61, and a 21% relative increase in death or MI, RR 1.21, 95% CI 0.99 to 1.47. These confidence intervals crossed 1.0, so the findings were not statistically significant by conventional thresholds.
Even so, from a clinical perspective, the point estimates are notable. In a high-risk AMI population with frequent hemodynamic instability, arrhythmia, ongoing ischemia, or multiorgan dysfunction, a trend toward worse outcomes with a restrictive threshold is difficult to dismiss, even when formal significance is not met.
Interaction testing
The key inferential question was whether the effect of transfusion strategy differed by ICU status. The interaction p values were 0.55 for death and 0.52 for death/MI. These values do not support statistically significant effect modification by critical illness status.
In practical terms, this means there is no evidence that critically ill patients behaved differently enough from noncritically ill patients to warrant a separate transfusion conclusion. The authors therefore conclude that the overall MINT results, which generally favor a liberal strategy, apply similarly across both settings.
Clinical Interpretation
This subanalysis reinforces an important clinical distinction: evidence supporting restrictive transfusion in broad ICU populations should not automatically be extrapolated to patients with active coronary ischemia. AMI creates a pathophysiologic environment in which oxygen delivery to vulnerable myocardium may be especially sensitive to reductions in hemoglobin concentration. The observation that restrictive transfusion did not perform better in either subgroup, and numerically performed worse in both, is consistent with that concept.
At the same time, the study should not be overread as definitive proof that every patient with AMI and anemia should be transfused to a hemoglobin target near 10 g/dL. The subgroup comparisons were not statistically significant, and subgroup analyses always carry less certainty than the parent trial result. Still, the absence of heterogeneity is informative. Clinicians should be cautious about using ICU admission as a reason to default to more restrictive transfusion in this population.
The results are particularly relevant for hospitals with restrictive blood management programs embedded in electronic order sets or stewardship pathways. Those programs are generally evidence-based, but this analysis suggests that AMI with anemia remains an exception where a more liberal threshold may be clinically justified. The finding also aligns with bedside experience: patients with myocardial ischemia often tolerate anemia poorly, especially when tachycardic, hypotensive, or requiring high oxygen demand.
Another practical message is that absolute risk matters. Critically ill patients had much higher event rates overall. Even if the relative treatment effect is similar across ICU and non-ICU settings, the potential absolute benefit of avoiding an overly restrictive strategy could be larger in critically ill patients because their baseline risk is greater.
Strengths and Limitations
Strengths
The analysis derives from a large, multicenter randomized trial with international enrollment, improving internal validity and broadening clinical relevance. The subgroup definition, ICU treatment at randomization, is straightforward and clinically recognizable. Endpoint selection was also appropriate, focusing on death and death/MI at 30 days, which are highly meaningful in AMI.
Limitations
Several limitations should shape interpretation. First, this was a subgroup analysis, and the trial was not primarily powered to detect treatment differences separately within critically ill and noncritically ill patients. Second, ICU status is a useful but imperfect marker of critical illness; thresholds for ICU admission vary across centers and countries. Third, the abstract does not provide detailed subgroup baseline characteristics, transfusion exposure, hemoglobin separation, infarct type, shock status, or invasive management patterns, all of which could influence outcome. Fourth, the confidence intervals are wide enough that moderate benefit or harm cannot be excluded with precision.
It is also important to remember that transfusion itself is not risk free. Potential adverse effects include volume overload, transfusion reactions, and immunologic or inflammatory consequences. The clinical question in AMI is therefore not whether transfusion is intrinsically beneficial, but whether maintaining a higher hemoglobin threshold better offsets ischemic risk than it adds transfusion-related risk. MINT suggests that, in AMI with anemia, the balance may tilt away from a very restrictive approach.
Implications for Practice and Research
For current practice, the most defensible takeaway is that clinicians should not assume restrictive transfusion is safer in AMI simply because a patient is in the ICU. If anything, the point estimates in critically ill patients lean in the opposite direction. In patients with AMI and anemia, especially those with ongoing ischemia, hemodynamic instability, or large infarcts, a liberal strategy remains a reasonable and evidence-supported consideration.
Future work should clarify whether specific AMI subgroups derive greater benefit from higher hemoglobin thresholds, including patients with ST-segment elevation MI, cardiogenic shock, severe left ventricular dysfunction, or recurrent ischemia. Analyses incorporating actual achieved hemoglobin levels, number of transfused units, timing of transfusion relative to ischemic events, and invasive versus conservative management would be especially helpful. Cost, blood utilization, and patient-centered outcomes such as functional recovery may also inform future policy.
Guideline panels will need to integrate MINT and related evidence carefully. The broader transfusion literature still strongly supports restrictive thresholds in many hospitalized adults, but AMI appears to be a clinically important exception or at least an area where equipoise is narrowing in favor of a more liberal threshold.
Funding and Trial Registration
The abstract for this subanalysis does not report funding details. The parent MINT trial was registered at ClinicalTrials.gov as NCT02981407. Readers should consult the full article and parent trial publication for detailed funding, oversight, and protocol information.
Conclusion
This MINT subanalysis shows that critically ill patients with AMI and anemia have worse short-term outcomes than noncritically ill patients, but there is no evidence that ICU status changes the relative effect of transfusion strategy. In both groups, restrictive transfusion was associated with numerically higher rates of death and death/MI than liberal transfusion, although differences were not statistically significant. The most clinically relevant message is that the broader MINT signal favoring a liberal strategy appears to extend across both critically ill and noncritically ill AMI populations.
References
1. Cooper HA, Fordyce CB, Herbert BM, Alexander JH, Rao SV, Brooks MM, Bertolet M, Carson JL, Hébert PC, Myocardial Ischemia and Transfusion (MINT) Trial Investigators. Red Cell Transfusion in Critically Ill and Noncritically Ill Patients With Acute Myocardial Infarction and Anemia-A Subanalysis of the Myocardial Ischemia and Transfusion (MINT) Trial. Critical Care Medicine. 2026-05-25. PMID: 42188994.
2. Carson JL, Brooks MM, Abbott JD, et al. Restrictive or Liberal Transfusion Strategy in Myocardial Infarction and Anemia. New England Journal of Medicine. 2023;389:2446-2456.
3. Hébert PC, Wells G, Blajchman MA, et al. A Multicenter, Randomized, Controlled Clinical Trial of Transfusion Requirements in Critical Care. New England Journal of Medicine. 1999;340:409-417.
4. Carson JL, Guyatt G, Heddle NM, et al. Clinical Practice Guidelines From the AABB: Red Blood Cell Transfusion Thresholds and Storage. JAMA. 2016;316:2025-2035.