Transfusion Thresholds and the MINT Trial: Beyond Journal Club with NEJM Group

Core IM Podcast

Time Stamps

• 01:29 History of Transfusion Guidelines/Studies
• 06:49 Pathophysiology of Anemia in heart disease
• 10:04 Transfusion in MI Studies
• 13:45 MINT Trial
• 22:20 Takeaways/Discussion

Sponsor:

Freed is an AI scribe that listens, prepares your notes, and writes your note in < 30 seconds. Freed learns your style over time and is HIPAA compliant!

Join over 9K clinicians who love Freed and saves them hours a day! Use the code CORE50 to get 50% off your first month with Freed.

Show Notes

Background

• Blood is a limited resource, costly, and not without harms, which include:
  • Transfusion reactions
  • Volume overload
  • Infection
  • Iron overload
  • Venous thromboembolism
• Guidelines
  • Historically, clinicians followed the 10/30 rule, which was to transfuse when hemoglobin dropped <10 or hematocrit <30
  • Most recent guidelines have recommended a more restrictive transfusion threshold of 7g/dL, but up to 8 g/dL for patients with cardiovascular disease
    • 2014 AHA/ACC guidelines: no benefit to routine transfusion for hemodynamically stable patients with NSTEMI and hemoglobin levels >8
    • 2023 AABB (Association for the Advancement of Blood & Biotherapies) guidelines: recommend a restrictive transfusion threshold of 7 g/dL (may consider a threshold of 8 g/dL for patients undergoing orthopedic surgery or have pre-existing cardiovascular disease)
• Pathophysiology: Why does hemoglobin matter?
  • Hemoglobin is responsible for delivering oxygen to organs
    • O2 delivery = CO x [1.39 x hemoglobin x SaO2 + (0.003 x PaO2)]
    • A JAMA study of healthy volunteers in 1998 showed that they could tolerate reduction of hemoglobin to 5 g/dL without evidence of inadequate oxygenation or abnormal lactate at rest
      • 2 participants developed ST changes with mild activity
        • Suggests ischemia provoked by low hemoglobin even in patients with normal coronaries
      • If this happens in healthy patients, what are the effects of anemia on a sicker hospitalized population?
        • In patients with MI, anemia may worsen myocardial ischemia, reducing the heart’s ability augment cardiac output to compensate for anemia, further worsening oxygen delivery

History of Transfusion Threshold Studies

• TRICC Trial (1999)
  • Population: Critically ill euvolemic patients with hemoglobin levels of <9g/dL
  • Treatment Groups:
    • Restrictive strategy: transfusion threshold of 7.0 g/dL
    • Liberal strategy: transfusion threshold of 10.0 g/dL
  • Sample Size: 838
  • Results: Overall 30-day mortality similar (18.7% in restrictive, 23.3% in liberal)
    • In-hospital mortality lower in restrictive group (22.2% vs 28.1%; p=0.05)
    • Subgroup analysis showed restrictive strategy may be beneficial in patients who were less sick (APACHE score of ≤20) or younger (<55 years of age)
• Circulation Study (2005)
  • Population: Patients enrolled in clinical trials of ACS (acute coronary syndrome)
  • Study Type: Cohort study looking at relationship between baseline hemoglobin levels and major adverse cardiovascular events
  • Sample Size: 39,922
  • Results:
    • In STEMI patients, cardiovascular mortality increased (OR 1.21) as hemoglobin levels fell below 14 g/dL
    • In NSTEMI patients, mortality increased (OR 1.45) as the hemoglobin fell below 11 g/dL
  • Caveats: Not a trial! This was observational data from heterogeneous trials; impossible to fully adjust for confounders
• Villanueva Trial (2013)
  • Population: patients with acute upper GI bleeding but not exsanguinating
  • Treatment Groups:
    • Restrictive strategy: transfusion threshold of 7 g/dL
    • Liberal strategy: transfusion threshold of 9 g/dL
  • Sample Size: 921
  • Results: Survival at 6 weeks was higher in the restrictive group (95%) vs liberal group (91%)
    • Liberal group had higher rates of further bleeding (16% vs 10%) and higher risk of adverse events (48% vs 40%).
      • The results were hypothesized to be due to blood transfusions increasingportal pressure, thus increasing the risk of re-bleeding
• These trials led to a paradigm shift in medicine
  • A restrictive hemoglobin transfusion threshold of 7 g/dL appeared to be better/noninferior in every population that was studied
  • This also included cardiac surgery patients
• However, no major trials had studied patients with heart attacks

Transfusion Threshold Studies in Patients with MI

• Transfusion in Symptomatic CAD (2013): first pilot study of transfusion thresholds in patients with ACS, done by Dr. Carson (the PI of the MINT trial)
  • Population: patients with ACS/stable angina and hemoglobin <1
  • Treatment Groups:
    • Restrictive strategy: transfusion threshold of 8 g/dL
    • Liberal strategy: transfusion threshold of 10 g/dL
  • Sample Size: 110
  • Results: 30-day composite of death, MI, or unscheduled revascularization was higher in the restrictive group (26% in restrictive group; 11% in liberal group)
    • 30-day all-cause death was higher in restrictive group (13%) vs 1.8% in the liberal group
  • Caveats:
    • Age was not matched between the two groups
    • Small study
• REALITY Clinical Trial (2021)
  • Population: patients in France and Spain with MI and hemoglobin between 7 and 10 g/dL
  • Treatment Groups:
    • Restrictive strategy: transfusion threshold of 8 g/dL
    • Liberal strategy: transfusion threshold of 10 g/dL
  • Sample Size: 668
  • Results: 30-day MACE was noninferior (11% in restrictive group; 14% in liberal group)
    • 30-day all-cause death was higher in liberal group (7.7% vs 5.6%).
  • Caveats:
    • Largest trial of patients with MI but still a relatively small trial
    • At a 1 year analysis, the restrictive strategy was no longer noninferior
    • RR for MACE was 1.13 and the 1-sided 97.5% CI was no longer within the prespecified noninferiority margin
• The MINT Trial
  • • Question: Does a liberal transfusion threshold improve outcomes in patients with acute MI?
      • Population: 3504 patients with myocardial infarction and a hemoglobin level of less than 10g/dL
        • Mean age 72.1 years; 45.5% were female
        • 1/3rd had history of MI, PCI or HF; half had CKD
        • 55.8% had type 2 MI; 41.7% had type 1 MI
        • 81% had NSTEMI
        • Included patients with type 1, 2, 4b, or 4c MI
          • Type 1 MI is due to acute plaque rupture/thrombosis
          • Type 2 MI is due to supply-demand mismatch due to a non-cardiac condition
          • Type 4b MI is secondary to stent/scaffold thrombosis after prior PCI
          • Type 4c MI is secondary to restenosis of stent/native coronary artery after balloon angioplasty
        • Exclusions:
          • Uncontrolled bleeding
          • On palliative treatment
          • Scheduled for cardiac surgery during the current admission
          • Declined to receive blood transfusion
      • Treatment Groups:
        • Restrictive strategy: transfusion threshold of 7 or 8 g/dL
          • Transfusion was permitted but not required when the hgb level was <8 and was strongly recommended when it was <7 or if patients had anginal symptoms not well-controlled with medication
        • Liberal strategy: transfusion threshold of 10 g/dL
        • Note: discontinuation of protocol was much higher in liberal group (13.7%) vs restrictive group (2.6%)
      • Results:
        • Primary outcome of 30-day myocardial infarction or all-cause death was similar: 16.9% in restrictive group; 14.5% in liberal group (p=0.07)
        • Secondary outcome of death at 30 days was 9.9% in restrictive and 8.3% in liberal (risk ratio 1.19, 95% CI 0.96-1.47)
          • Risk of cardiac death was significantly higher in the restrictive vs liberal group (5.5% vs 3.2%).
        • Composite secondary outcome at 30 days (death, MI, revascularization, rehospitalization) was 19.6% in restrictive and 17.4% in liberal (risk ratio 1.13, 95% CI 0.98-1.29)
        • Risk of TACO (transfusion-associated cardiac overload) was lower in restrictive group (0.5% vs 1.3%; RR 0.35; 95% CI 0.16-0.78)
        • Risk of HF at 30 days was similar, as was risk of PE/DVT and transfusion reactions
      • Subgroup Analyses: remember that subgroup analyses are hypothesis-generating, not hypothesis-testing
        • Among patients with Type 1 MI; restrictive strategy had significantly more primary outcome events than liberal (18.2% vs 13.8%)
          • No significant difference was seen in patients with Type 2 MI
        • Most subgroup analyses favored the liberal strategy but particularly:
          • Patients with eGFR <30 mL/min not on dialysis (RR 1.36; 1.02-1.83)
          • Patients with NSTEMI (RR 1.19; 1.00-1.41)
      • Limitations:
        • Primary outcome did not technically reach significance – p was 0.07
          • Maybe this was because the event rate was lower than expected
            • The observed difference was 15% and the trial had been powered to detect a 20% relative difference
        • ~35% of participants each group had received a transfusion prior to enrollment (on average, 2-3 units)
        • 5x more patients discontinued the protocol in the liberal group
        • The treatment groups were not masked from providers or patients
          • The restrictive group did not receive sham transfusions
            • Placebo control more important when using subjective endpoints like pain or quality of life
          • Could anemia have affected decisions like revascularization?
            • No difference in PCI or coronary catheterization rates between the groups MINT Trial (2023)

Discussion

• Technically a neutral trial but with a confidence interval that barely crosses 1 and a p-value that barely misses 0.05
  • One more patient reaching the endpoint in the restricted arm would have made the p-value 0.05 (the findings are ‘fragile’)
• A letter to the editor argued that from a Bayesian perspective, even if you went in with a skeptical pretrial probability, there is a high probability of harm from a restrictive strategy in the MI population
• Why enroll patients with type 2 MI?
  • More generalizability
    • There are more hospitalized patients with type 2 MI than type 1
  • Easier to reach enrollment goals
  • Caveat: patients with type 2 MI are very heterogeneous
  • The potential benefit of liberal transfusions seen in type 1 patients was not seen in type 2 patients
    • Reminder: also a subgroup analysis, so not conclusive
• UpToDate now cites the MINT trial, saying that patients with acute MI and risk factors for reinfarction may benefit from transfusion if their hemoglobin <10
• Why was there a 5x higher rate of protocol discontinuation in the liberal group?
  • Many listed reasons including:
    • Side effects: volume overload, transfusion reactions
    • Patient or provider preference
    • Blood supply shortage; staffing issues
  • This affects real world applicability – blood transfusion are not benign and reinforces that we should not blindly follow a protocol just to get to a hemoglobin target

Takeaways

• A subset of patients having acute MI may benefit from a higher hemoglobin goal of 10 mg/dL compared to standard of 7-8 g/dL
  • Patients with Type 1 MI may have more benefit
  • We should try to figure out who is having worsening ischemia from their anemia booking for ischemic symptoms (chest pain, ST changes)
  • We should also tailor treatment to the patient in front of us:
    • Balance risk factors from transfusion (volume overload, transfusion reactions)
• Many questions remain about how and if this will change guidelines and implications for the blood supply
  • Per the MINT authors, treating all acute MI patients in the US with anemia with a liberal transfusion strategy would require 360,000 more units of RBCs compared to a restrictive strategy
    • Extrapolating their results, adopting a liberal transfusion strategy could prevent 4800 patients from dying or having recurrent myocardial infarction

Transcript

Dr. Gregory Katz: Welcome to the second season of Beyond Journal Club, a collaboration between Core IM and NEJM Group.

Dr. Clem Lee: The goal of Beyond Journal Club is to take landmark clinical trials and put them into context, telling the story of how we got to where we are and what it means for how we take care of our patients.

Dr. Suellen Li: I’m Dr. Suellen Li, a former editorial fellow at the New England Journal of Medicine and a hospitalist at MGH.

Dr. Gregory Katz: I’m Dr. Greg Katz, a cardiologist at NYU.

Dr. Clem Lee: I’m Dr. Clem Lee, also a former fellow and a current guest editor at the New England Journal of Medicine.

Dr. Suellen Li: Today we’re talking about the MINT trial, which stands for “Myocardial Ischemia and Transfusion.” MINT was published in the New England Journal of Medicine in November of 2023.

Dr. Clem Lee: MINT was a trial looking at transfusion thresholds in patients with anemia and acute myocardial infarction – what we will refer to as MI.

Dr. Gregory Katz: When MINT first came out, my initial thought was just a surprise. Surprise that someone thought that this trial was worth doing, and then even more surprise about the results.

Dr. Clem Lee: So today to help us understand the MINT trial in a bit, we will first get into the story of what the prior transfusion trials told us.

Dr. Gregory Katz: Then, we’ll get into the pathophysiology of anemia and why it might be uniquely harmful in the setting of an acute MI.

Dr. Suellen Li: And finally we’ll deep dive into the MINT trial, to try to answer the question of whether patients who are having heart attacks really should have a different threshold for blood transfusions.

History of Transfusion Guidelines/Studies

Dr. Suellen Li: For a long time, the standard for transfusion was something called the 10/30 rule, which was to transfuse for a hemoglobin of 10 or hematocrit of 30. So where does the dogma that we write in our notes all the time “transfuse to Hgb of 7 g/dl” come from?

Dr. Gregory Katz: A lot of this perspective landmark TRICC trial, which was published in the NEJM in 1999. This was a trial that randomized critically ill patients in the ICU to transfuse when Hgb drops below 7 aka a restrictive or transfuse when Hgb drops below 10 aka a liberal transfusion strategies and compared their 30 day all-cause mortality.

Dr. Clem Lee:The TRICC Trial was groundbreaking and changed practice for many doctors. It showed that 30-day mortality was similar in the two groups

Dr. Gregory Katz: And when you dig into the details, it seems that a restrictive strategy wasn’t just neutral, but there was a suggestion transfusing at lower hemoglobin thresholds was better than the liberal strategy.

Dr. Suellen Li: Yep in the subgroup analysis of TRICC, those with cardiac disease had similar mortality whether transfused with a liberal or restrictive strategy. So The TRICC investigators concluded that critically ill patients should be transfused for a hemoglobin threshold of 7 but because they did not enroll patients with active coronary ischemia, they noted that population might be an exception.

Dr. Gregory Katz: Over the next two decades, there was study after study of transfusion thresholds in many different populations that essentially all found the same thing – restrictive thresholds are non-inferior and potentially even superior to liberal transfusion thresholds

Dr. Clem Lee: Better outcomes when transfusing at lower Hgb threshold was even replicated in patients undergoing cardiac surgery and even in patients who were having active GI bleeds.

Dr. Gregory Katz: Yeah those results of GI bleeding study were impressive, and for me, totally counterintuitive. The idea of a restrictive strategy in an acute GI bleed for me, is totally counterintuitive.

Dr. Suellen Li: Yep the Villanueva Trial, which was published in the NEJM in 2013. This randomized patients with severe acute upper GI bleeding to either a restrictive transfusion threshold of 7 g/dL or liberal threshold of 9 g/dL.

Dr. Gregory Katz: I would have thought that giving more blood to people who were bleeding would improve outcomes. But the restrictive strategy was associated with a 45% reduction in all-cause mortality and also reduced the rate of rebleeding.

Dr. Clem Lee: It turns out that more blood transfusions increases portal pressure and thus increases the risk of re-bleeding. So this was pretty convincing evidence that a lower hemoglobin threshold of 7 was appropriate for even patients with acute UGIB.
Dr. Gregory Katz: So the consistency of the transfusion data led to a complete paradigm shift for clinical medicine. And it makes sense why restrictive strategies might be noninferior or better if you think about it – blood transfusion is not a benign procedure.

Dr. Clem Lee: Anyone who’s had to consent patients for a blood transfusion knows these risks well – infection, transfusion reactions, volume overload, thrombosis, the formation of new antibodies, and graft-versus-host disease.

Dr. Suellen Li: That reminds me of a patient of mine who was chronically anemic with a baseline hemoglobin around 7. On the day of discharge, he had a hemoglobin of 6.9. We went back and forth and decided to give him a unit of red blood cells to “tank him up” before he left and chased it with some IV furosemide. That same evening – not even the next day – I got a notification that he had been re-admitted for volume overload.

Dr. Clem Lee: Oof those re-admission notifications always gives that sinking feeling in my stomach – like wait did i cause that readmission? Did I do any harm?

Dr. Gregory Katz: We are all doing our best in these gray areas but it is blood transfusions come with real harms and now we have multiple studies that demonstrate that more isn’t always better and it may even be worse.

Dr. Clem Lee: So to recap the trials that we had in the world of pRBC blood transfusions thresholds- we started with the TRICC trial which surprised people initially that lower thresholds may be ok in critically ill patients, then tons of other studies including in the GI world with the Villanueva study showed that lower thresholds were equal or better

Dr. Suellen Li: Yep that’s the same information, Dr. Jeff Carson, the lead investigator of the MINT trial was going off of.

Jeff Carson: I think everyone was leaning towards a restrictive transfusion approach across all groups of patients. And for wasn’t unreasonable given that all the trials trial after trial after trial. So if you look at all the major trials that’ve been published in England, England Journal of Medicine, and every one of those trials said less blood is a safe. So the presumption was, well, that’s going to be the same in patients with underlying myocardial infarction.

Dr. Clem Lee: Yep so the assumption was that patients with myocardial infarctions would also benefit from transfusing only when a hgb reached a lower, restrictive threshold but we didnt know for sure and did not have that level of evidence

Dr. Gregory Katz: Yeah -I’m not sure its right that we can extrapolate the data that we have on ICU patients, patients with cardiac surgery, and patients having GI bleeds to a group of patients having heart attacks. And so I think it might be worth taking take a quick look at the pathophysiology of heart attacks and why anemia in acute MI may be a little different in anemia in other settings, than anemia in other settings.

Pathophysiology of Anemia in heart disease

Dr. Gregory Katz: If you think about it from a pathophysiology perspective, one could make a good argument for why we should think about transfusing heart attacks patients differently than other patients.

Dr. Clem Lee: Of course the cardiologist is going to try to convince us that the heart is more special than the other organs.

Dr. Gregory Katz: Well it is! So in a heart attack these muscles aren’t getting blood flow but the thing that makes a heart attack different from other patient populations is that the primary problem in heart attack is lack of O2 delivery to heart muscle. That’s different from sepsis or GI bleed or severe hyponatremia -its just a different process. The follow up to that – if the anemia is making that O2 delivery worse, then it is going to increase the size of the MI and lead to higher risk of heart failure, mechanical complications, arrhythmias, RAAS activation. There’s a million places where the anemia is making that primary problem worse and maybe a patient with ischemic myocardium really doesn’t fit into all of the other patient populations.

Dr. Clem Lee: Well is there clinical data to back up that pathophysiology or are you just biased about the heart? You know it made theoretical sense to give blood in GI bleed but that didnt pan out in the Villanueva study.

Dr. Suellen Li: So actually yes, we do have real data. In 1998, there was a study published in JAMA of 32 healthy, young volunteers where the researchers took out blood and replaced it with albumin or plasma until they reached a hemoglobin of 5 g/dL from a baseline of about 13.

Dr. Gregory Katz: Suellen I appreciate you supporting me despite Clem’s skepticism. What is shocking, besides the fact that this got past an IRB, is that at rest, the patients with that Hgb of 5 manifested no evidence of inadequate oxygenation or perfusion. They all had normal oxygen and lactate levels.

Dr. Suellen Li: But here is the kicker: two patients with normal coronaries in that study developed ST changes with activity

Dr. Gregory Katz: The two patients with ST changes tell me that young healthy people can have ischemia provoked by low hemoglobin. So that means people with significant enough anemia can “fail a stress test” even if their coronaries are normal.

Dr. Suellen Li: But most of our patients in the hospital are not healthy volunteers. And when you think about patients with heart disease, even milder anemia could have potentially serious myocardial consequences.

Dr. Gregory Katz: Right so just to say it explicitly, if we think of an MI: there’s an occlusion of the coronary artery that is decreasing blood flow to the heart. You throw in some anemia and it could make the myocardial ischemia that much worse.

Dr. Suellen Li: So, the anemia here is like a snowball effect but in a bad direction – anemia increases myocardial ischemia, which reduces the heart’s ability to compensate for a low hemoglobin, which further worsens oxygen delivery, and the process just spirals.

Dr. Gregory Katz: We cardiologists might be biased about the heart being special but this is the reason why guidelines have historically recommended a higher transfusion threshold in patients with cardiovascular disease – 8 compared to 7.

Dr. Clem Lee: Not to be a troll again – The theoretical pathophys doesn’t matter if the trials don’t back up that hypothesis – so what have the data shown and not shown in the world of transfusion in MIs?

Dr. Gregory Katz: As they say – the road to hell is paved with biological plausibility.

Transfusion in MI Studies

Dr. Gregory Katz: As we were talking about before, patients having active MIs have historically been excluded from transfusion trials. And so even though we have theories about the pathophysiology, until recently we haven’t had any hard outcomes data to give us a sense of how these patients do.

Dr. Suellen Li: So the seeds for the MINT trial were planted more than two decades ago, before the TRICC trial was even published.

Dr. Clem Lee: Which is pretty impressive, considering I plant mint seeds every summer and they never last past the fall for me…

Dr. Suellen Li: Well I think we’re all glad you’re a doctor and not a farmer, Clem.

Dr. Gregory Katz: I agree. That was a good professional decision.

Dr. Suellen Li: We talked to Dr. Jeffrey Carson about where he got the idea for looking into transfusions in the active MI population and why he has been thinking about this for two decades.

Dr. Jeffery Carson: We had done a long time ago a study in patients of the Jehovah’s Witness faith in which we looked at 2000 or so of those who underwent an operation in the operating room. It’s an observational study. And what we found was an interaction with cardiovascular disease that is patients were more likely to die as their hemoglobin fell in patients with cardiovascular disease than those without cardiovascular disease. And that mortality stepped up around the often criticized 10 gram threshold. So we had some observational data that this group might be at risk.

Dr. Suellen Li: So in these surgical patients with cardiac disease who declined blood transfusion, he actually found that the lower the preop hgb level, the higher their operative mortality

Dr. Clem Lee: And so this led to several other observational studies in patients with acute coronary syndrome.

Dr. Suellen Li: For example, one study in Circulation in 2005, there was a signal that patients who were having STEMIs with Hgb levels below 14 have a higher mortality and likewise in patients with NSTEMI, a hgb below 11 had a higher CV mortality

Dr. Gregory Katz: I’ll try to stay off of my full observational data soapbox, but let’s just remind everyone that a cohort study never proves anything causal – it just provides data to help develop a hypothesis to test in a randomized trial.

Dr. Suellen Li: Yes, so in the symptomatic coronary artery disease population, randomized trial data were essentially absent, that was until Dr. Carson and his colleagues performed a pilot trial in 2013.

Dr. Jeffrey Carson: We did a pilot study also funded by NIHA number of years ago where we enrolled 110 patients in that pilot. And lo and behold, we had seven deaths in the lower group in the restrictive group and one death in the liver group. So like holy mackerels, maybe we have really something going on here

Dr. Suellen Li: So now we had more data that maybe cardiac patients were different and a liberal approach to blood transfusions could be better in the heart disease population.

Dr. Clem Lee: And then came the first large, major randomized trial, the REALITY trial published in JAMA in 2021. The researchers of REALITY randomized 668 patients with MI and anemia to a restrictive threshold of 8 or liberal threshold of 10 g/dL.

Dr. Gregory Katz: The results of REALITY were actually consistent with the prior randomized data on transfusion and NOT consistent with the observational data we had on MI patients – so they showed a restrictive threshold to transfuse to a hgb goal of 8 was noninferior to a liberal threshold to 10.

Dr. Clem Lee: oh so maybe the REALITY trial gave us a reality check

Dr. Suellen Li: But it’s hard to think of a small trial like REALITY with under 700 patients as the last word on the topic.

Dr. Clem Lee: And that brings us to the MINT trial.

MINT Trial

Dr. Suellen Li: So the MINT trial randomized 3504 patients (that’s 5 times as many patients as in the REALITY trial!) with myocardial infarction and a hemoglobin <10 to either a restrictive or liberal transfusion threshold.

Dr. Clem Lee: And these patients feel similar to the patients I commonly see the mean age of these patients was about 72 years and 45% of the patients were female. About half had CKD and a third had a history of HF, PCI or MI

Dr. Gregory Katz: I’ll also point out that the vast majority of these patents with MI, 81% to be exact, had an NSTEMI.

Dr. Suellen Li: Yep so the MINT investigators took these patients who presented with these different types of acute myocardial infarction and randomized them to a liberal transfusion threshold of 10 g/dl or to a restrictive threshold to transfuse to a hemoglobin of either 7 or 8 g/dl.

Dr. Gregory Katz: That threshold of 7 or 8 seems confusing – so let’s dig into the details here. At a hemoglobin of 8, transfusion was permitted but not required. At a hemoglobin of 7, it was strongly recommended. It was also strongly recommended to transfuse if patients had angina that was not well-controlled with medication.

Dr. Suellen Li: And then in the liberal group, all the patients were transfused one unit at a time to achieve a hemoglobin level of 10 g/dL.

Dr. Gregory Katz: So by day 1, the liberal strategy group had an average hemoglobin of 10.1 and the restrictive group had an average hemoglobin level of 8.8.

Dr. Suellen Li: The primary outcome was MI or death at 30 days.

Dr. Gregory Katz: Suellen, do you want to have the honor of the results…

Dr. Suellen Li: Yes, If you want to be literal about the results of the trial, it was neutral.

Dr. Clem Lee: But if you look into it, surprisingly an MI or death occurred more in the restrictive transfusion arm — and to be exact 17% of patients in the restrictive transfusion group had a death or MI vs. 15% of patients in the liberal group — giving us a p-value of 0.07.

Dr. Gregory Katz: So we have a confidence interval that barely crosses 1 with a p value just over the magical 0.05 that we look for. But even though we tend to be strict about our p values, I often come back to the idea that our p value cutoff is somewhat arbitrary – I mean, is 0.07 truly different than 0.05?

Dr. Suellen Li: Rather than just get into an existential argument about the nature of p values, why don’t we dig into the results in more detail so that we can actually understand what’s happening rather than just a top line result?

Dr. Clem Lee: So 4 big things stand out to me. First, these interventions were not blinded — the restrictive group didn’t get sham transfusions.

Dr. Gregory Katz: Right yes when a trial is unblinded I would be cautious about looking at soft outcomes that are subject to the placebo effect. But the primary outcome in MINT was a composite of centrally adjudicated MI and cardiac death – neither of which are really subject to the placebo effect.

Dr. Clem Lee: Ok fair, second thing that made me pause was a possible confounder. MI patients with anemia may be less likely to receive aggressive antithrombotic treatment or go for invasive cardiac procedures, so did that impact the MINT results?

Dr. Suellen Li: I doubt it because the rates of PCI and cath were similar between the two groups.

Dr. Gregory Katz: But I do like where you’re going with thinking how anemia interacts with medical therapy for coronary disease.It can be a big confounder in the observational data. In other words, did the patient with a lower hemoglobin do worse because they were anemic and suffered the effects of end-organ hypoxemia or was it because the anemia made us less likely to give them DAPT and take them to the cath lab?

Dr. Suellen Li: I wondered about that too. And the third thing to consider is the side effects in each arm. The rates of protocol discontinuation were much higher in the liberal group compared to the restrictive group – there was a 5-fold difference.

Dr. Gregory Katz: It’s really notable to me that almost 14% of patients assigned to the liberal transfusion strategy discontinued this protocol. That’s more than 1 in 8 patients.

Dr. Suellen Li: Right that’s a lot. And when we think of why it’s because some in the liberal group discontinued because of adverse effects like fluid overload or transfusion reactions

Dr. Clem Lee: Unsurprisingly, the rates of TACO (transfusion-associated circulatory overload) were lower in the restrictive group compared to the liberal group. But oddly the rates of heart failure were similar.

Dr. Gregory Katz: TACO bout a surprising finding. No but in all seriousness, others discontinued transfusion protocol in the liberal arm because blood supply shortages or patient or clinician preference.

Dr. Suellen Li: I mean, we’ve been told for years that a restrictive strategy is better so I see why some clinicians might balk at the idea of transfusing patients to a hemoglobin of 10.

Dr. Clem Lee: I agree – I would’ve also hesitated. Ok so the last and maybe most relevant point to me as a generalist is that a majority of these patients had a type 2 MI – 56%, whereas only 42% had a type 1 MI.

Dr. Suellen Li: The different terms for MIs kind of get thrown around and can get confusing, so let’s just take a second to discuss what it means to have an MI?

Dr. Gregory Katz: So For the universal definition of myocardial infarction, you need to have means troponin + symptoms/EKG/echo. So a bit of troponin leak does NOT mean a myocardial infarction. You need something else clinical to make the case.

Dr. Clem Lee: But that universal definition of MI is kind of like an catch all term/umbrella. It doesn’t tell us anything about which type of MI it is.

Dr. Suellen Li: Yep the mint investigators included most types of MIs, so patients with both type 1 and type 2 MIs, and also type 4b and 4c.

Dr. Gregory Katz: And just to be on the same page, type 1 refers to coronary occlusion from ruptured atherosclerotic plaque and type 2 is more related to supply/demand mismatch. Type 4b and 4c are MIs related to stent thrombosis or in stent restenosis.

Dr. Clem Lee: And the type of MI also has nothing to do with whether the ST segments are up or not.

Dr. Suellen Li: With that in mind, if you look at the subgroup that had a type 1 MI aka coronary occlusion, the liberal strategy to transfuse to hgb of 10 was actually superior than to restrict these patients to a hgb of 7-8.

Dr. Gregory Katz: If you think about it, patients with Type 1 MI are very different from patients with Type 2 MI. The type 2 MI group is far more heterogeneous – what proportion of all-comers to the ICU don’t have shortness of breath or chest pain and a troponin elevation? I mean, you could even go so far as to argue that these type 2 MI patients are essentially just slightly less sick TRICC trial patients, meaning all-comers who are sick and in the hospital.

Dr. Clem Lee: Right, it makes you wonder whether, if this trial only enrolled Type 1 MI patients, the overall primary endpoint may have been significantly different between the groups without all the heterogeneity of the type 2 MI patients.

Dr. Suellen Li: So we sat down with Dr. Sunil Rao, director of interventional cardiology at NYU and one of the investigators on the MINT trial to ask why they enrolled patients with both Type 1 and Type 2 MI.

Dr. Sunil Rao: We felt that type two MI is an important understudied population, obviously anemia has been implicated as a cause of type two mi. From a practical trialist standpoint, is not a traditional industry funded drug A versus drug B, acute coronary syndrome trial. These patients were really sick. If you think about the patients in the hospital now who are hospitalized with acute MI, who have anemia, think about all their other comorbid conditions, they tend to be much, much older. They have a lot of end stage renal disease or CKD. These are patients that traditionally are not approached for ACS trials, right? So it is challenging and it was challenging to enroll in this trial. So we felt that if we could broaden the inclusion criteria, it would make the trial a little bit more feasible to enroll in.”

Dr. Suellen Li: That’s fair. We all do see a lot of patients with type 2 MIs and it would be helpful to have a sense of whether they benefit from a higher transfusion threshold.

Dr. Clem Lee: So the investigators did their best to make this a pragmatic trial that would attract as many patients to enroll as possible. Unfortunately, they overestimated how many MIs the patients were going to have – so the fact that there were fewer MIs may be one explanation of why the p value was 0.07 and not 0.05.

Dr. Suellen Li: So let’s summarize the MINT Trial. At face value, this was a neutral trial. there was no difference in death or MI at 1 month after transfusing to a hgb goal of 7 or 8 vs. 10 for patients with MIs and anemia. But, the p-value barely missed significance and the confidence interval barely crossed 1 so there’s an argument to be made that this was actually a positive trial.

Dr. Clem Lee: And the most interesting to me was the subgroup analysis. Patients with type 1 MI did do better with liberal strategy transfusing to Hgb of 10.

Takeaways/Discussion

Dr. Suellen Li: So where does this leave us? Will the MINT trial change your practice? Are you going to be more liberal with your transfusion threshold for patients with MI?

Dr. Clem Lee: I think there’s two questions there. The first is a standalone question about whether that p value of 0.07 is a problem for us and should we consider this a positive trial.

Dr. Suellen Li: So one of the statisticians even did a calculation that just a single additional patient reaching the endpoint in the restrictive arm would have made the p-value 0.05.

Dr. Clem Lee: But then again, on the other hand if there was one more event in the liberal group that would have made it more neutral!

Dr. Gregory Katz: And also, remember when you’re thinking about applying a liberal transfusion strategy that almost 14% of patients in the liberal group didn’t ultimately follow the protocol to transfuse to hgb of 10 – if those patients had followed the protocol, the results could have been very different- we don’t really know.

Dr. Suellen Li: It’s a reminder of just how fragile some trial results can be, especially when there are low event rates across relatively small groups of patients.

Dr. Clem Lee: Exactly, that leads to the second question which how to put the results of the MINT trial in the context of the body of literature – which until now basically favored restricting blood transfusions to a lower threshold

Dr. Gregory Katz: The degree to which any trial persuades us to change our practice is going to be necessarily based on our pre-trial expectations. – it’s no different than using the concept of a pre-test probability when we interpret a diagnostic study. Most doctors incorporate their pre-trial expectation into the way we interpret and integrate a new trial into our practice.

Dr. Clem Lee: What Greg is referring to is called a Bayesian analysis — and what your pretrial probability was and in this case, how skeptical you were about liberal transfusions – ultimately affects how you interpret the results of the MINT trial.

Dr. Gregory Katz: This type of analysis was in a letter to the editors – a Bayeseian analysis to understand how the results of MINT would influence your thinking based on your pre-trial probability. Everyone I talked about this trial brought up that letter to the editor.the authors of that letter suggested that even if you went in with a skeptical pretrial probability that liberal transfusions would be beneficial, when you analyze the results through a Bayesian lens there’s still a strong chance that the results of MINT should tell us that liberal transfusion is actually beneficial for an MI population

Dr. Clem Lee: Really? But all the other trials showed that liberal wasn’t better and this trial technically on face value was neutral too. So putting that all together, how would this one trial persuade us to be more liberal about transfusions?

Sunil Rao: The MINT trial wasn’t as definitive as we would have hoped, but I do think that if you look at it from the 10,000 foot view, from a practical clinician point of view, strong consideration should be given to transfusing, that type one MI patient to a hemoglobin of 10, because again, all the point estimates favored that strategy. The cardiac death favored that strategy. I was very, very conservative with transfusion because of all the data to date. But now I think with the Mint trial, which is the largest randomized trial ever done in this population, is 3,500 patients. It really, I think, does suggest and note, I’m not saying it proves. It suggests that there is a benefit with more aggressive transfusion.

Dr. Suellen Li: You know I have to say I think this is what makes clinical research imperfect but also so interesting. You have this very thoughtfully designed and conducted trial and yet at the end of the day, different people may interpret the results in different ways. What do you guys think we are MINT to take away from this trial?

Dr. Clem Lee: Other than the fact that perhaps we shouldn’t be letting you attempt to write puns…

Dr. Suellen Li: Ouch! haha Clem I’m trying!

Dr. Clem Lee: Touche – yeah I do think I’ll be more thoughtful about my patients who are having an MI when I’m thinking about whether to transfuse to a goal of 7 or 10.

Dr. Suellen Li: You know, I actually saw that after the MINT trial was published UpToDate now recommends considering transfusing to a Hgb of 10 in patients with acute MI.

Dr. Clem Lee: Yeah I saw that too. They did not specify the type of MI but say to consider it in patients with “unstable MI or stable MI with risk factors … such as incomplete revascularization or high-risk coronary anatomy”

Dr. Suellen Li: When we asked Dr. Carson, he said that in his view, a hemoglobin of 7-8 should be avoided for all patients with acute MIs and so basically transfuse if a patient with an MI has a Hgb of for example 7.5.

Dr. Gregory Katz: You know, I understand that perspective, but I don’t agree with it.

Dr. Suellen Li: You don’t?

Dr. Gregory Katz: Yeah I don’t think that we should be transfusing every anemic patient with an acute MI. But I do think it’s fair to say that I ultimately find the data from MINT persuades me that the idea that there is a subset of patients with acute MIs are doing worse because of their anemia and we should correct that problem with transfusion.

Dr. Clem Lee: So we should not be transfusing every patient with an MI who has anemia to a hemoglobin goal of 10. We individualize our threshold for each patient. It’s a bloody difficult task but this trial is bringing us closer to an answer.

Dr. Gregory Katz: One thing we have an answer for is that none of you should be making puns for the rest of this episode… but yeah our job is to figure out who is ischemic because of their anemia – are they having chest pain, do they have ST depressions – that’s the patient who may need a higher transfusion target. Our job as doctors is to figure out which of our patients are doing worse because of their anemia and thus will benefit from being more liberal with our approach to blood transfusions.

Dr. Suellen Li: That point about treating the patient in front of you is so important – so if your patient develops volume overload or a febrile transfusion reaction, you don’t need to blindly transfuse to a target of 10 just because the MINT trial suggested that you should.

Dr. Clem Lee: I too am curious to see how the MINT results end up changing clinical practice and guidelines and to see whether the transfusion pendulum will begin to swing.

Dr. Suellen Li: Yeah This trial has so many implications – if we were to shift practice for example what would that mean for the blood supply and shortages?

Sunil Rao: Now, I’m not too nervous about it because remember, it took the British Navy 200 years to put limes on their ships to prevent scurvy. So we’re still within that 200 year timeframe. I’m hopeful that practice will change a little faster than that, though.”

Dr. Suellen Li: Well I would certainly hope so! And on that note, it’s a wrap on this episode of Beyond Journal Club!

Dr. Clem Lee: If you found this episode helpful, please share it with your team and colleagues and give it a rating on Apple Podcasts or whatever podcast app you use. It really does help people find us.

Dr. Gregory Katz: And if you have any feedback, please email us at hello@coreIMpodcast com. Opinions expressed are our own and do not represent the opinions of any affiliated institutions.

References

• Amsterdam EA, Wenger NK, Brindis RG, Casey DE Jr, Ganiats TG, Holmes DR Jr, Jaffe AS, Jneid H, Kelly RF, Kontos MC, Levine GN, Liebson PR, Mukherjee D, Peterson ED, Sabatine MS, Smalling RW, Zieman SJ. 2014 AHA/ACC Guideline for the Management of Patients with Non-ST-Elevation Acute Coronary Syndromes: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2014 Dec 23;64(24):e139-e228.
• Carson JL, Stanworth SJ, Guyatt G, Valentine S, Dennis J, Bakhtary S, Cohn CS, Dubon A, Grossman BJ, Gupta GK, Hess AS, Jacobson JL, Kaplan LJ, Lin Y, Metcalf RA, Murphy CH, Pavenski K, Prochaska MT, Raval JS, Salazar E, Saifee NH, Tobian AAR, So-Osman C, Waters J, Wood EM, Zantek ND, Pagano MB. Red Blood Cell Transfusion: 2023 AABB International Guidelines. JAMA. 2023 Nov 21;330(19):1892-1902.
• Weiskopf RB, Viele MK, Feiner J, Kelley S, Lieberman J, Noorani M, Leung JM, Fisher DM, Murray WR, Toy P, Moore MA. Human cardiovascular and metabolic response to acute, severe isovolemic anemia. JAMA. 1998 Jan 21;279(3):217-21.
• Hébert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G, Tweeddale M, Schweitzer I, Yetisir E. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med. 1999 Feb 11;340(6):409-17.
• Sabatine MS, Morrow DA, Giugliano RP, Burton PB, Murphy SA, McCabe CH, Gibson CM, Braunwald E. Association of hemoglobin levels with clinical outcomes in acute coronary syndromes. Circulation. 2005 Apr 26;111(16):2042-9.
• Villanueva C, Colomo A, Bosch A, Concepción M, Hernandez-Gea V, Aracil C, Graupera I, Poca M, Alvarez-Urturi C, Gordillo J, Guarner-Argente C, Santaló M, Muñiz E, Guarner C. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013 Jan 3;368(1):11-21.
• Mazer CD, Whitlock RP, Fergusson DA, Hall J, Belley-Cote E, Connolly K, Khanykin B, Gregory AJ, de Médicis É, McGuinness S, Royse A, Carrier FM, Young PJ, Villar JC, Grocott HP, Seeberger MD, Fremes S, Lellouche F, Syed S, Byrne K, Bagshaw SM, Hwang NC, Mehta C, Painter TW, Royse C, Verma S, Hare GMT, Cohen A, Thorpe KE, Jüni P, Shehata N; TRICS Investigators and Perioperative Anesthesia Clinical Trials Group. Restrictive or Liberal Red-Cell Transfusion for Cardiac Surgery. N Engl J Med. 2017 Nov 30;377(22):2133-2144.
• Carson JL, Brooks MM, Abbott JD, Chaitman B, Kelsey SF, Triulzi DJ, Srinivas V, Menegus MA, Marroquin OC, Rao SV, Noveck H, Passano E, Hardison RM, Smitherman T, Vagaonescu T, Wimmer NJ, Williams DO. Liberal versus restrictive transfusion thresholds for patients with symptomatic coronary artery disease. Am Heart J. 2013 Jun;165(6):964-971.e1.
• Ducrocq G, Gonzalez-Juanatey JR, Puymirat E, Lemesle G, Cachanado M, Durand-Zaleski I, Arnaiz JA, Martínez-Sellés M, Silvain J, Ariza-Solé A, Ferrari E, Calvo G, Danchin N, Avendaño-Solá C, Frenkiel J, Rousseau A, Vicaut E, Simon T, Steg PG; REALITY Investigators. Effect of a Restrictive vs Liberal Blood Transfusion Strategy on Major Cardiovascular Events Among Patients With Acute Myocardial Infarction and Anemia: The REALITY Randomized Clinical Trial. JAMA. 2021 Feb 9;325(6):552-560.
• Gonzalez-Juanatey JR, Lemesle G, Puymirat E, Ducrocq G, Cachanado M, Arnaiz JA, Martínez-Sellés M, Silvain J, Ariza-Solé A, Ferrari E, Calvo G, Danchin N, Avendano-Solá C, Rousseau A, Vicaut E, Gonzalez-Ferrero T, Steg PG, Simon T; REALITY Investigators. One-Year Major Cardiovascular Events After Restrictive Versus Liberal Blood Transfusion Strategy in Patients With Acute Myocardial Infarction and Anemia: The REALITY Randomized Trial. Circulation. 2022 Feb 8;145(6):486-488.
• Carson JL, Brooks MM, Hébert PC, Goodman SG, Bertolet M, Glynn SA, Chaitman BR, Simon T, Lopes RD, Goldsweig AM, DeFilippis AP, Abbott JD, Potter BJ, Carrier FM, Rao SV, Cooper HA, Ghafghazi S, Fergusson DA, Kostis WJ, Noveck H, Kim S, Tessalee M, Ducrocq G, de Barros E Silva PGM, Triulzi DJ, Alsweiler C, Menegus MA, Neary JD, Uhl L, Strom JB, Fordyce CB, Ferrari E, Silvain J, Wood FO, Daneault B, Polonsky TS, Senaratne M, Puymirat E, Bouleti C, Lattuca B, White HD, Kelsey SF, Steg PG, Alexander JH; MINT Investigators. Restrictive or Liberal Transfusion Strategy in Myocardial Infarction and Anemia. N Engl J Med. 2023 Dec 28;389(26):2446-2456.
• Khan MS, Spertus JA, Chan PS. Transfusion Strategy in Myocardial Infarction and Anemia. N Engl J Med. 2024 Mar 7;390(10):960-961.
• UpToDate. Overview of the acute management of non-ST-elevation acute coronary syndromes.

The post Transfusion Thresholds and the MINT Trial: Beyond Journal Club with NEJM Group appeared first on Core IM Podcast.