Highlight
– In 1,424 high–cardiac-risk veterans who developed postoperative anemia after major vascular or general surgery, a liberal transfusion strategy (trigger Hb <10 g/dL) did not reduce the composite of 90‑day death or major ischemic events compared with a restrictive strategy (trigger Hb <7 g/dL).
– The primary outcome occurred in 9.1% of patients in the liberal group vs 10.1% in the restrictive group (RR 0.90; 95% CI, 0.65–1.24).
– A prespecified secondary outcome—cardiac complications without myocardial infarction—was lower with liberal transfusion (5.9% vs 9.9%; RR 0.59; 99% CI, 0.36–0.98), suggesting possible reduction in arrhythmias, heart failure, or nonfatal cardiac arrest but requiring cautious interpretation.
Background: the clinical question and why it matters
Perioperative anemia is common after major surgery and is associated with worse outcomes. Red blood cell (RBC) transfusion can rapidly raise hemoglobin and improve oxygen delivery but carries risks including transfusion reactions, volume overload, and immunomodulation, as well as resource implications. Over two decades of trials in critical care and elective surgery have supported restrictive transfusion strategies (commonly Hb thresholds near 7 g/dL) for many patient populations, reducing transfusions without increasing mortality. However, whether restrictive thresholds are safe in patients at particular risk for myocardial ischemia after major operations—patients in whom diminished oxygen carrying capacity could precipitate myocardial injury—has been an ongoing clinical uncertainty. Current guideline statements acknowledge uncertainty for patients with active cardiac ischemia and for major cardiac risk populations, leaving practice variation across perioperative and vascular surgery care.
Study design: the TOP randomized clinical trial
TOP was a parallel-group, single-blind, randomized superiority trial conducted across 16 Veterans Affairs Medical Centers in the United States. Between February 2018 and March 2023, investigators enrolled 1,428 veterans aged 18 years or older who were at high cardiac risk and underwent major vascular or general surgery. After developing postoperative anemia (hemoglobin <10 g/dL), 714 participants were randomized to a liberal transfusion strategy with a trigger of Hb <10 g/dL and 714 to a restrictive strategy with a trigger of Hb <7 g/dL. The trial was single-blind (participants blinded), and care teams followed assigned transfusion thresholds. The primary endpoint was a composite of all-cause death, myocardial infarction, coronary revascularization, acute kidney injury, or ischemic stroke within 90 days after randomization. Secondary endpoints included a composite of cardiac complications other than myocardial infarction (arrhythmias, heart failure, nonfatal cardiac arrest) among others.
Population and intervention details
Of 1,424 patients included in the analysis, mean age was 69.9 years (SD 7.9); the cohort was predominantly male (97.8%), reflecting the VA population. Racial distribution included 18.8% Black, 4.1% Hispanic, and 75.2% White patients. The majority (91.1%) underwent vascular surgical procedures. Hemoglobin separation between strategies was modest but clinically meaningful: mean difference of 2.0 g/dL on day 5 after randomization.
Key findings
Primary outcome: The composite primary outcome occurred in 9.1% (61 of 670) of patients in the liberal group versus 10.1% (71 of 700) in the restrictive group (relative risk [RR] 0.90; 95% CI, 0.65–1.24). The confidence interval crosses unity and the point estimate does not support a statistically significant superiority of the liberal strategy for the primary composite endpoint.
Secondary outcomes: Among the prespecified secondary end points, the composite of cardiac complications without myocardial infarction occurred less frequently in the liberal group: 5.9% (38 of 647) vs 9.9% (67 of 678) in the restrictive group (RR 0.59; 99% CI, 0.36–0.98). This result suggests fewer arrhythmias, episodes of heart failure, or nonfatal cardiac arrest with a liberal transfusion threshold.
Other outcomes and safety: The trial report notes no overall increase in major adverse ischemic events or mortality with the restrictive strategy. Data on transfusion reactions, infection rates, volume overload/heart failure events, length of stay, and resource use were collected and discussed in the full manuscript; the summary results did not demonstrate a clear signal of harm from either strategy apart from the secondary cardiac composite favoring liberal transfusion. The modest hemoglobin separation (≈2 g/dL by day 5) underscores that the clinical difference in oxygen delivery between groups was limited to the early postoperative period when risk may be highest.
Interpretation: what the TOP results mean for clinicians
The TOP trial provides high-quality randomized data specific to a high–cardiac-risk surgical population—predominantly older male veterans undergoing vascular surgery—addressing a practical and controversial perioperative decision. The primary take-home is that liberal postoperative transfusion to a hemoglobin threshold of <10 g/dL did not reduce the composite of death or major ischemic events at 90 days compared with a restrictive strategy using a <7 g/dL trigger.
However, the observed reduction in non‑MI cardiac complications with the liberal threshold merits careful consideration. Arrhythmias and heart-failure events can drive morbidity, readmission, and resource consumption; if a true benefit exists, it could meaningfully affect postoperative care. That said, this was a secondary endpoint, and despite statistical significance using a conservative CI (99% CI reported), the finding should be considered hypothesis‑generating until replicated or supported by meta-analysis and subgroup analyses that identify which patients are most likely to benefit.
How this fits with existing evidence and guidelines
The TOP results align with a body of trials supporting restrictive transfusion strategies in many settings. Key prior studies include the TRICC trial in critically ill patients (NEJM 1999), which established noninferiority of a restrictive threshold (~7 g/dL) in the ICU; the FOCUS trial in elderly patients undergoing hip surgery (NEJM 2011), which found no mortality benefit to liberal transfusion; and randomized data in acute upper gastrointestinal bleeding showing benefit for restrictive transfusion in selected patients (Villanueva et al., NEJM 2013). The AABB guidelines have recommended restrictive transfusion thresholds (Hb ~7–8 g/dL) for most hospitalized stable patients, while acknowledging uncertainty for those with active cardiac ischemia or high ischemic risk.
TOP provides direct evidence in a high‑cardiac‑risk, postoperative surgery cohort. Its neutral primary result supports continued use of restrictive thresholds as a safe default in many postoperative patients, but the secondary finding raises the possibility that some cardiac complications could be mitigated by higher Hb targets in select patients.
Limitations and considerations
– Generalizability: The study population was predominantly male veterans (≈98% male) and largely underwent vascular surgery (91%), which may limit extrapolation to women, other surgical populations (e.g., major abdominal or oncologic procedures in civilians), and to centers with different transfusion practices.
– Single‑blind design: Treating clinicians were unblinded to transfusion assignment, which could influence clinical decision‑making and secondary outcomes such as arrhythmia detection or heart‑failure diagnosis.
– Magnitude of Hb separation: The achieved mean hemoglobin difference was modest (2.0 g/dL by day 5). Whether a larger separation or different transfusion timing would change outcomes remains unknown.
– Secondary outcome interpretation: The cardiac composite without MI was a secondary endpoint; although statistically significant with a 99% CI, it requires confirmatory study. Multiplicity and post hoc explanations remain a concern.
– Event definition and adjudication: The validity of composite endpoints depends on robust event adjudication; full methods and adjudication processes are reported in the primary paper and should be reviewed when applying findings.
Practical clinical recommendations
– For most patients at high cardiac risk after major vascular or general surgery who develop postoperative anemia, a restrictive transfusion strategy (trigger Hb <7 g/dL) remains reasonable and did not increase 90‑day death or major ischemic events in TOP.
– Consider individualization: In patients with ongoing ischemia, refractory angina, hemodynamic instability, or worsening heart failure/arrhythmias that may plausibly be related to anemia, individualized transfusion to higher hemoglobin targets remains defensible pending further evidence.
– Monitor for cardiac complications: Given the signal favoring fewer non‑MI cardiac events with liberal transfusion, clinicians should remain vigilant for arrhythmias and heart‑failure signs in anemic postoperative patients and consider multidisciplinary discussion (surgery, cardiology, anesthesia/intensive care) when contemplating transfusion above restrictive thresholds.
Research and policy implications
– Need for replication: Confirmation of the secondary cardiac composite finding in other surgical populations and in trials that include more women and non‑veteran patients is required before changing broad practice recommendations.
– Mechanistic studies: Investigation into whether transfusion reduces arrhythmogenic substrate or prevents ischemia‑triggered heart failure postoperatively would help explain the observed signal.
– Resource stewardship: Transfusion is costly and not without risk. Findings supporting restrictive transfusion preserve blood bank resources and reduce exposure to transfusion risks when safe.
Conclusion
The TOP randomized trial provides important evidence specific to a high–cardiac-risk postoperative population: liberal transfusion (trigger Hb <10 g/dL) did not lower 90‑day all‑cause death or major ischemic events compared with a restrictive threshold (Hb <7 g/dL). A reduction in non‑MI cardiac complications with the liberal strategy invites further study but should not yet prompt wholesale abandonment of restrictive thresholds. Clinicians should individualize transfusion decisions, balancing ischemic risk, symptoms, physiologic reserve, and transfusion risks while following evolving evidence and guideline updates.
Funding and trial registration
The TOP trial was supported and conducted across Veterans Affairs Medical Centers. Trial registration: ClinicalTrials.gov Identifier: NCT03229941. Full funding and disclosure details are reported in the published JAMA article.
References
1. Kougias P, Sharath SE, Zhan M, Carson JL, Norman LE, Mi Z, et al.; TOP Trial Investigators. Liberal or Restrictive Postoperative Transfusion in Patients at High Cardiac Risk: The TOP Randomized Clinical Trial. JAMA. 2025 Nov 8:e2520841. doi:10.1001/jama.2025.20841. PMID: 41205227; PMCID: PMC12596742.
2. Hébert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G, et al.; Transfusion Requirements in Critical Care Investigators for the Canadian Critical Care Trials Group. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med. 1999;340(6):409–417.
3. Carson JL, Terrin ML, Noveck H, Sanders DW, Chaitman BR, Rhoads GG, et al.; FOCUS Investigators. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med. 2011;365(26):2453–2462.
4. Villanueva C, Colomo A, Bosch A, Concepción M, Hernandez-Gea V, Aracil C, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013;368(1):11–21.
5. Carson JL, Guyatt G, Heddle NM, Grossman BJ, Cohn CS, Fung MK, et al. Clinical practice guidelines from the AABB: red blood cell transfusion thresholds and storage. Ann Intern Med. 2016;165(3):192–213. (AABB clinical practice guideline)
Author note
This article synthesizes the TOP trial findings and places them into clinical context for perioperative clinicians, cardiologists, and surgeons involved in transfusion decision-making. Readers are encouraged to consult the full JAMA publication for detailed methods, subgroup analyses, and supplementary data.
