Perfusion Tools

Weekly evidence briefing · Abstract-based

Weekly CPB & ECMO Research

New CPB and ECMO research, briefed for perfusion practice: what the study found, why it matters, what to watch at the pump, and where the evidence is still weak.

November 2–8, 2026

This week's signal: one newly published study met our relevance and date criteria. Rather than padding the list with older papers that only appeared in a November issue, this briefing focuses on the strongest new clinical signal: criteria-based ECMO use during lung transplantation and early neurologic complications.
  1. Lung transplant · ECMO/CPBRetrospective cohortTransplantationNovember 2
    Priority: HighClinical usefulness: 4.2/5

    Criteria-based ECMO during lung transplantation was associated with fewer ischemic strokes

    Study: Incidence and Risk Factors for Early Major Neurologic Complications in Lung Transplantation: From Restrictive Cardiopulmonary Bypass to ECMO by Criteria.

    Design: Single-center retrospective cohort of 389 lung transplant recipients, comparing an earlier restrictive-CPB era (2012–2017; n=288) with a later criteria-based ECMO era (2022–2024; n=101).

    Key findings

    • Early major neurologic complications occurred in 10% versus 5% of patients; the difference in the proportion of affected patients was not statistically significant (P=0.152).
    • Total neurologic events were substantially more frequent in the earlier cohort (78 vs 8 events; incidence rate ratio 3.5, 95% CI 1.7–7.4).
    • Ischemic stroke was lower with criteria-based ECMO use: 2% versus 9% (P=0.022).
    • Intraoperative mechanical circulatory support was used more often in the ECMO-era cohort (43% vs 20%).
    • The apparent neurologic benefit came with trade-offs: greater intraoperative blood loss (2489 vs 1861 mL) and more severe primary graft dysfunction at 24 and 72 hours.
    Perfusionist briefing: This does not mean ECMO should routinely replace CPB for every lung transplant. The useful signal is that a predefined, criteria-based support strategy may reduce exposure to unstable hypoxemia and cerebral hypoperfusion while avoiding CPB when it is not required. If ECMO is chosen, cerebral oxygenation, upper-body gas exchange, cannula position and flow, anticoagulation, blood loss, and graft reperfusion all remain part of the same safety problem.

    Limitations: The two groups came from different eras, so surgical technique, donor/recipient selection, ICU care, transfusion practice, and ECMO experience may all have changed. The later cohort was also older and had more cardiopulmonary comorbidity. This is hypothesis-generating evidence, not proof that ECMO itself caused the lower stroke rate.

    Read the PubMed record (PMID 42968367) →

October 26 – November 1, 2026
Week in review: the strongest perfusion signal was mechanistic rather than outcome-changing. HERACLES directly compared IABP and a microaxial flow pump during VA-ECMO, while the ECPR studies reinforced that post-reperfusion physiology and center-level systems matter. None of these studies establishes a new device-selection or prognostication threshold.
  1. VA-ECMO · LV unloadingRandomized mechanistic trialCirc Cardiovasc IntervOctober 31
    Priority: HighClinical usefulness: 4.5/5

    HERACLES: IABP improved myocardial oxygen supply–demand balance during VA-ECMO

    Study: Hemodynamic Effects of Reducing LV Afterload With Impella or Intra-Aortic Balloon Counterpulsation During Veno-Arterial Extracorporeal Membrane Oxygenation in Cardiogenic Shock: HERACLES Trial.

    Design: Randomized mechanistic trial of 24 patients with cardiogenic shock on VA-ECMO, comparing IABP with a microaxial flow pump. Most patients were SCAI stage D/E; median LVEF was 10%.

    Key findings

    • The myocardial oxygen supply–demand ratio improved with IABP but not with the microaxial pump: 1.48 versus 0.97 (P=0.036).
    • The difference was driven mainly by reduced LV systolic pressure-time integral; both devices reduced LV pressures.
    • Coronary flow increased marginally with IABP and was unchanged with the microaxial pump.
    • Thirty-day mortality was similar (38% vs 45%; P=0.84); the trial was not designed to prove a survival difference.
    Perfusionist briefing: “LV unloading” is not a single physiological effect. Two devices can both lower LV pressure while producing different coronary-flow and myocardial oxygen-balance responses. For VA-ECMO patients, device choice should still be tied to the actual problem—LV distension, pulmonary edema, absent aortic opening, myocardial recovery, or systemic perfusion—not to this small surrogate-endpoint trial alone.

    Limitations: Only 24 patients were randomized. The primary endpoint was mechanistic, not mortality or recovery, and invasive measurements were made under controlled study conditions. The findings do not establish clinical superiority of IABP over a microaxial pump.

    Read the PubMed record (PMID 42966011) →
  2. ECPR · ReperfusionRetrospective cohortASAIO JournalOctober 29
    Priority: HighClinical usefulness: 4.2/5

    After ECPR, the 12-hour lactate response carried more information than the prereperfusion value

    Study: Dynamic Early Postreperfusion Physiology Is Associated With Organ Recovery After Extracorporeal Cardiopulmonary Resuscitation.

    Design: Retrospective study of 91 patients with refractory out-of-hospital cardiac arrest treated with ECPR; paired prereperfusion and 12-hour lactate values were available in 38 patients.

    Key findings

    • Twelve-hour lactate was lower in survivors than nonsurvivors (5.45 vs 9.04 mmol/L; P=0.002), while prereperfusion lactate did not differ.
    • In paired data, mortality discrimination increased from AUC 0.47 before reperfusion to 0.80 at 12 hours.
    • Each 5 mmol/L higher 12-hour lactate was associated with higher mortality (OR 2.73), lower odds of ECMO weaning (OR 0.44), and more cerebral edema/hypoxic injury (OR 2.07) after adjustment for low-flow duration.
    Perfusionist briefing: The useful message is the trajectory, not a single lactate cutoff. After ECPR, falling lactate should be interpreted alongside ECMO flow, native pulsatility, MAP, Hb, gas exchange, vasopressor dose, urine output, and regional perfusion. A persistently high lactate at 12 hours should prompt reassessment of unresolved shock or ischemia, but it is not a stand-alone futility rule.

    Limitations: Small retrospective cohort, only 38 paired measurements, and strong potential for confounding by shock severity and treatment response. The study identifies associations and does not validate a treatment threshold.

    Read the PubMed record (PMID 42956360) →
  3. MI cardiogenic shock · ECMONational observational cohortJ Am Heart AssocOctober 30
    Priority: Moderate–HighClinical usefulness: 4.0/5

    ECMO in infarction-related cardiogenic shock: large dataset, but no valid causal comparison

    Study: National Use Patterns and Comparative Effectiveness of Extracorporeal Membrane Oxygenation in Myocardial Infarction-Related Cardiogenic Shock.

    Design: Medicare-based retrospective cohort of 102,718 patients with MI-related cardiogenic shock from 2016–2023: ECMO 1,165, Impella 10,064, IABP 24,631, and medical management 66,858.

    Key findings

    • ECMO use increased from 0.8% to 1.8%, while the proportion of hospitals using ECMO rose from 9.2% to 17.1%.
    • Unadjusted 30-day mortality was 69% with ECMO, 60% with Impella, 39% with IABP, and 46% with medical therapy.
    • IPTW and G-formula analyses still estimated ECMO mortality at 68–70%, but a grace-period analysis yielded 54% mortality for both ECMO and Impella.
    • The authors explicitly concluded that valid causal inference about ECMO and survival could not be established.
    Perfusionist briefing: Do not interpret the crude mortality table as “ECMO is worse.” VA-ECMO is commonly deployed later and in more profound shock. The study is more useful as a reminder that cannulation timing, arrest status, shock stage, and prior MCS must be captured when comparing outcomes across programs.

    Limitations: Administrative Medicare data, an older population, residual confounding by indication, and no detailed information on ECMO flow, cannulation, LV unloading, lactate, or shock phenotype.

    Read the PubMed record (PMID 42960751) →
  4. Pediatric ECPRELSO registryCritical Care MedicineOctober 29
    Priority: Moderate–HighClinical usefulness: 3.9/5

    Pediatric ECPR outcomes still varied between centers after risk adjustment

    Study: Variation in Pediatric Extracorporeal Cardiopulmonary Resuscitation Survival After In-Hospital Cardiac Arrest: Development and Validation of a Survival Model for Risk-Standardized Survival Rates.

    Design: ELSO Registry multicenter cohort of 1,377 children with in-hospital cardiac arrest treated with ECPR (2020–2023); derivation n=1,060 and internal validation n=317.

    Key findings

    • The 14-variable model had AUC 0.728 in derivation and 0.678 in validation, with good internal calibration.
    • After patient-level risk adjustment, between-hospital variation remained significant; median odds ratio was 1.30 (95% CI 1.11–1.93).
    • The authors cautioned that discrimination is not strong enough for the model to be used as a stand-alone bedside decision tool.
    Perfusionist briefing: This is a systems paper as much as a prognostication paper. Differences in ECPR activation, cannulation readiness, low-flow time, team training, and post-cannulation management may contribute to center-level variation. It supports auditing ECPR processes, not using a score to deny an individual child support.

    Limitations: Internal validation only, registry-variable constraints, possible residual differences in case mix and center practice, and only fair discrimination for individual prediction.

    Read the PubMed record (PMID 42951894) →
  5. Cardiac surgery · CPBRandomized double-blind trialJ Am Heart AssocOctober 30
    Priority: LowClinical usefulness: 3.0/5

    Stellate ganglion block did not prevent postoperative atrial fibrillation after valve surgery

    Study: Efficacy of Stellate Ganglion Block for Preventing Postoperative Atrial Fibrillation After Cardiac Surgery: A Randomized, Double-Blind, Placebo-Controlled Trial.

    Design: Single-center randomized double-blind trial of 140 valve-surgery patients undergoing CPB; 132 were included in the modified intention-to-treat analysis.

    Key findings

    • POAF occurred in 43% after stellate ganglion block versus 40% with saline (adjusted P=0.989): the primary endpoint was negative.
    • Secondary signals favored SGB for VF/VT requiring defibrillation after cross-clamp release (6% vs 19%), postoperative hs-troponin I, and AKI (5% vs 18%).
    Perfusionist briefing: The defibrillation-after-unclamping signal is interesting for the operating room, but this is not a perfusion intervention and should not change current CPB practice. It is best viewed as a hypothesis about autonomic modulation and reperfusion arrhythmias.

    Limitations: Single-center trial, negative primary endpoint, and the favorable arrhythmia, troponin, and AKI findings were secondary outcomes that require replication.

    Read the PubMed record (PMID 42960736) →

October 19–25, 2026
Week in review: two studies were especially relevant to perfusion systems: a nationwide target-trial emulation of ECPR timing and a temperature-strategy study in emergency aortic surgery. Both support physiology- and system-aware decision making, but neither supplies a new hard cutoff for cannulation time or hypothermia temperature.
  1. ECPR · TimingTarget-trial emulationEHJ Acute Cardiovascular CareOctober 25
    Priority: TopClinical usefulness: 4.7/5

    Earlier ECPR was associated with the largest benefit, but 60 minutes was not an absolute futility boundary

    Study: Earlier Versus Later Extracorporeal Cardiopulmonary Resuscitation in Refractory Out-of-Hospital Cardiac Arrest: A Nationwide Sequential Landmark Target Trial Emulation.

    Design: Korean nationwide OHCA registry study (2016–2023) using sequential landmark target-trial emulation. Among 58,668 refractory OHCA patients, 909 received ECPR.

    Key findings

    • Weighted 30-day survival with ECPR versus continued conventional CPR was 31.0% vs 16.0% at the 15-minute landmark, 21.6% vs 7.2% at 30 minutes, and 10.9% vs 3.2% at 60 minutes.
    • Adjusted survival differences were +15.0, +14.4, and +7.7 percentage points, respectively (all P<0.001).
    • Favorable neurologic outcomes followed a similar pattern; exploratory benefit decreased with longer door-to-ECMO time but remained statistically significant through 120 minutes.
    Perfusionist briefing: The operational message is “earlier is better,” not “late is useless.” For an ECPR program, the modifiable targets are recognition, team activation, cannulation readiness, and time to full flow. A patient crossing 60 minutes should not be rejected solely because of the clock if the rest of the candidacy profile remains favorable.

    Limitations: Observational data despite sophisticated target-trial methods, residual confounding and selection remain possible, and the cohort was restricted to presumed cardiac-cause refractory OHCA. The results do not prove benefit for every late ECPR candidate.

    Read the PubMed record (PMID 42926135) →
  2. Aortic surgery · CPBRetrospective cohortPerfusionOctober 22
    Priority: HighClinical usefulness: 4.4/5

    Emergency hemiarch surgery: warmer hypothermia strategies showed a stroke signal, not a new temperature target

    Study: Impact of hypothermia level on neurological and renal outcomes in emergency aortic surgery: a single-centre experience.

    Design: Single-center retrospective cohort of 175 acute aortic syndrome operations (2018–2023); 89% underwent hemiarch replacement. Patients were grouped as ≤20°C, 20.1–24°C, 24.1–28°C, and >28°C.

    Key findings

    • CPB duration fell progressively with warmer strategies, from a median 270.5 minutes in deep hypothermia to 164 minutes in mild hypothermia.
    • Overall 30-day mortality was 18.9% and stroke 19.4%.
    • In the hemiarch subgroup, stroke occurred in 27%, 26%, 8.8%, and 11% across the four temperature groups (P=0.049).
    • The four-category association was no longer significant after multivariable adjustment; a separate >24°C versus ≤24°C sensitivity analysis was associated with lower stroke risk (OR 0.23, 95% CI 0.08–0.64).
    • No clear AKI advantage was demonstrated across temperature groups.
    Perfusionist briefing: This supports the direction of modern hemiarch practice—avoiding unnecessarily deep cooling when an appropriate cerebral-protection strategy is used—but it does not establish “24°C” as a universal target. Procedure extent, ACP strategy, arrest time, rewarming, embolic risk, and institutional experience still determine the temperature plan.

    Limitations: Retrospective, single center, major coupling between temperature, cerebral-perfusion strategy, CPB duration, procedure complexity, and era. The mild-hypothermia group was small, and results for total arch replacement remain uncertain.

    Read the PubMed record (PMID 42911832) →
  3. CPB · Neurologic safetyRetrospective MRI studyJ Cereb Blood Flow MetabOctober 20
    Priority: Moderate–HighClinical usefulness: 3.8/5

    Cerebral microbleed burden tracked with extracorporeal-circulation exposure

    Study: Cerebral microbleed burden after aortic and cardiac surgery is associated with extracorporeal circulation exposure: Evidence from young patients with heritable connective tissue disorders.

    Design: Retrospective brain-MRI study of 111 genetically confirmed patients with heritable connective tissue disorders, relating cerebral microbleeds to surgical and extracorporeal-circulation exposure.

    Key findings

    • All 58 patients exposed to surgery using extracorporeal circulation had cerebral microbleeds, compared with 4 of 53 surgery-naïve patients.
    • Microbleeds were predominantly lobar.
    • Burden increased with repeated procedures and, when duration data were available, with longer extracorporeal-circulation exposure (P<0.001).
    • Valve type and surgical technique were not associated with microbleed burden.
    Perfusionist briefing: This is a strong imaging signal that neurologic injury after extracorporeal circulation may be cumulative and multifactorial. It reinforces meticulous de-airing, minimizing avoidable circuit exposure and suction-related air/blood interfaces, and attention to embolic and perfusion safety—but it does not identify a single responsible device or technique.

    Limitations: Highly selected connective-tissue-disorder population, retrospective imaging, no direct measurement of intraoperative embolic load, and no demonstrated link between microbleed count and functional neurologic outcomes.

    Read the PubMed record (PMID 42899414) →
  4. Heart transplant · VA-ECMORetrospective cohortBiomedicinesOctober 21
    Priority: ModerateClinical usefulness: 3.6/5

    Axillary graft versus direct femoral arterial return after heart transplant: no proven superiority

    Study: Axillary Dacron Graft Versus Direct Femoral Arterial Cannulation for Peripheral VA-ECMO After Heart Transplantation: A Single-Center Cohort Study.

    Design: Retrospective single-center cohort of 42 heart-transplant recipients requiring peripheral VA-ECMO for primary graft dysfunction: axillary Dacron graft n=17 versus direct femoral arterial cannulation n=25.

    Key findings

    • Cannulation-related complications occurred in 0/17 axillary cases and 3/25 femoral cases (12%); OR 0.18, 95% CI 0.01–3.80, P=0.260.
    • Successful weaning was similar: 94.1% versus 92.0%.
    • Survival to discharge was 76.5% versus 64.0% (P=0.505), and stroke 5.9% versus 16.0% (P=0.632).
    • None of the between-group differences reached statistical significance.
    Perfusionist briefing: Axillary return offers antegrade systemic flow and avoids direct femoral return, but this small series cannot prove fewer vascular or neurologic complications. Cannulation choice should remain anatomy- and urgency-specific, with attention to achievable flow, limb/upper-extremity perfusion, graft hemostasis, and differential oxygenation.

    Limitations: Very small single-center cohort, wide confidence intervals, nonrandom cannulation selection, and limited power for uncommon vascular or neurologic complications.

    Read the PubMed record (PMID 42940214) →
  5. ECMO bridge · Lung transplantRetrospective cohortPneumologieOctober 21
    Priority: ModerateClinical usefulness: 3.3/5

    Age ≥50 years was associated with lower long-term graft survival after ECMO bridging

    Study: Age-related Long-Term Outcomes after ECMO Bridging to Lung Transplantation: A 15-Year Single-Center Experience.

    Design: Fifteen-year single-center retrospective cohort of 115 adults listed for lung transplantation while receiving ECMO; 93 underwent transplantation. Patients were prespecified as <50 versus ≥50 years.

    Key findings

    • Post-transplant 1-, 3-, and 5-year graft survival was 87%, 83%, and 77% in recipients <50 years versus 72%, 53%, and 44% in those ≥50.
    • Age ≥50 was independently associated with graft loss (adjusted HR 2.53, 95% CI 1.14–5.59; P=0.022).
    • Among follow-up survivors, older recipients had more comorbidity, lower exercise capacity, and lower return-to-work rates, while marked frailty remained uncommon.
    Perfusionist briefing: This is primarily a candidate-selection and recovery study, not an ECMO-management study. Age should not be treated as a stand-alone exclusion threshold; functional reserve, frailty, organ dysfunction, rehabilitation potential, and trajectory on ECMO remain more informative at the bedside.

    Limitations: Single-center experience spanning 15 years of evolving ECMO and transplant practice, substantial selection before listing and transplantation, and limited generalizability to all ECMO bridge candidates.

    Read the PubMed record (PMID 42906575) →

Selected for relevance to CPB and ECMO practice, not as a complete list of each week's publications. These summaries are based on abstracts and are for education; they are not treatment recommendations.