I now have excellent data from multiple sources. Let me compile the comprehensive comparison.
Sevoflurane vs. Ketamine vs. Propofol in HFrEF Anesthesia: Head-to-Head Evidence Review
Important framing: No head-to-head RCT has compared all three agents specifically in HFrEF/reduced-EF patients. The evidence is assembled from: (1) pharmacological studies in general populations, (2) cardiac surgery trials (CABG, valve), (3) specific HFrEF/low-EF studies, and (4) physiologic extrapolation. The closest direct comparison in low-EF patients is the propofol-ketamine vs. etomidate-midazolam study in CABG with EF <40%.
Head-to-Head Hemodynamic Profile in HFrEF
| Parameter | Sevoflurane | Propofol infusion | Ketamine infusion |
|---|
| Cardiac output/index | Mild-moderate decrease (dose-dependent) | -15% to -21% (CI and SI) | -21% in catecholamine-depleted HF (RCT) |
| Systolic BP | Decreases (vasodilation + ↓ SVR) | -25% to -40% (induction bolus) | +13% MAP (but from increased SVR) |
| SVR | Decreases (↓ afterload - favorable in HFrEF) | Decreases (↓ afterload - favorable) | +38% (very unfavorable in HFrEF) |
| HR | Minimal change / mild reflex tachycardia | Minimal change; may blunt baroreflex | Increases (sympathomimetic) |
| PCWP / filling pressures | Decreases | Decreases | +20% (worsens congestion) |
| mPAP | Slight decrease | Slight decrease | +14% (dangerous in biventricular failure) |
| Myocardial O₂ demand | Decreases | Decreases | Increases |
| Net effect on failing LV | Moderate depression; reduced afterload partially offsets; cardioprotective properties | Significant depression + hypotension risk; least direct cardioprotection | Afterload increase overwhelms sympathomimetic benefit; direct myocardial depression unmasked |
1. SEVOFLURANE
Hemodynamic Effects
Sevoflurane causes dose-dependent myocardial depression and systemic vasodilation. In HFrEF, the SVR reduction is actually a relative advantage - the failing LV is afterload-sensitive, and lowering SVR can improve cardiac output (Barash Clinical Anesthesia, 9e). Dose titration with MAC monitoring is feasible and allows careful control.
Cardioprotective Properties (Key Differentiator)
Sevoflurane triggers anesthetic preconditioning - a mechanism that mimics ischemic preconditioning via:
- Diffusion through myocardial cell membranes
- Mitochondrial electron transport alteration → reactive oxygen species generation
- Protein kinase C activation → KATP channel opening
- ~30-40% of cardioprotection related to reduced calcium loading during ischemia
This is highly relevant in HFrEF where the myocardium is chronically ischemic/hibernating. Multiple meta-analyses found volatile anesthetics reduced myocardial damage in cardiac surgery. However, the 2019 MYRIAD trial (pragmatic multicenter RCT) found no significant reduction in 1-year mortality or myocardial ischemia with volatile agents in elective CABG - the cardioprotective benefit is not consistently reproducible in modern clinical settings. (Barash Clinical Anesthesia, 9e; Miller's Anesthesia 10e, p.2482)
The Jiao et al. 2019 meta-analysis (PMID: 31661512; 89 RCTs, 14,387 CABG patients) found no significant difference in operative mortality (RR 0.92, CI 0.68-1.24) or 1-year mortality between volatile anesthetics and TIVA in CABG. ICU stay was shorter with volatiles, but Trial Sequential Analysis showed evidence remains inconclusive for most outcomes.
The Guinot et al. 2020 RCT (PMID: 33327246) found sevoflurane did not reduce myocardial injury (cTnI) vs. propofol in cardiac surgery, and sevoflurane was associated with higher incidence of acute renal failure and higher GDF-15 (inflammatory stress marker).
The HypnoRenalRIP 2025 RCT (PMID: 41093687) found sevoflurane - but not propofol - preserved remote ischemic preconditioning's renoprotective effect in cardiac surgery, suggesting propofol attenuates conditioning signals that sevoflurane preserves.
Summary for HFrEF
Sevoflurane is the most physiologically compatible volatile agent for HFrEF when used at low MAC concentrations (0.5-1.0 MAC). Its SVR-lowering effect is appropriate, its cardioprotective mechanisms are mechanistically compelling (even if clinical translation is inconsistent), and it does not raise pulmonary pressures or afterload. Requires careful titration to avoid excessive myocardial depression at high doses.
2. PROPOFOL INFUSION (TIVA)
Hemodynamic Effects (from Miller's Anesthesia 10e, p.2479-2481)
Propofol causes the most pronounced hemodynamic disruption at induction:
- -25% to -40% systolic BP from a 2.0-2.5 mg/kg induction dose
- Cardiac index -15%, stroke volume index -20%, LVSWI -30%
- SVR -15 to -25% (vasodilation - favorable for LV afterload, but simultaneous CO depression is net negative)
- Blunts the baroreflex - the tachycardic response to hypotension is attenuated, a dangerous feature in HFrEF where HR compensation is critical
- Hemodynamic depression lags the hypnotic effect by 2-7 minutes (effect-site equilibration) - blood pressure drops further after the patient loses consciousness
During infusion (maintenance), hemodynamic effects are much less than induction bolus, and the myocardial oxygen supply/demand ratio is preserved (reduces both MBF and O₂ consumption proportionally).
Propofol Infusion Syndrome (PRIS) Risk
Prolonged high-dose propofol infusion (>48h at >5 mg/kg/h) can cause PRIS: dysrhythmias, acute heart failure, hyperkalemia, metabolic acidosis, rhabdomyolysis, and lipemia. In a compromised myocardium (HFrEF), the cardiac components of PRIS are particularly dangerous. This limits use in ICU-level prolonged sedation. - Tintinalli's Emergency Medicine
Cardioprotective Properties
Propofol is an antioxidant (phenolic structure similar to vitamin E) and may reduce free radical-induced reperfusion injury. However, Miller's Anesthesia (10e) and clinical trial data show propofol is less cardioprotective than volatile agents - two large studies comparing propofol with sevoflurane in cardiac surgery showed lower troponin levels and better hemodynamic function with sevoflurane. Propofol also appears to attenuate remote ischemic preconditioning signaling (HypnoRenalRIP 2025 RCT).
However, at very high doses (120 mcg/kg/min), propofol may have dose-dependent cardioprotection. Combinations of volatile preconditioning followed by propofol post-conditioning act synergistically - a potentially useful hybrid strategy.
Summary for HFrEF
Propofol induction is risky in HFrEF due to the profound, baroreflex-blunted hypotensive response. When used, dose reduction (0.5-1.0 mg/kg), slow titration, and vasopressor support are essential. For maintenance infusion, propofol TIVA at titrated rates is a reasonable option if the patient's BP and CO are supported. Prolonged high-dose infusion should be avoided (PRIS risk).
3. KETAMINE INFUSION
(Prior session provides detailed evidence; summary here for direct comparison)
Hemodynamic Effects in HFrEF - The Critical Evidence
Christ et al. 1997 RCT (PMID: 9209606) - 25 ICU patients with HF, mean LVEF 30 ± 7%, randomized to ketamine infusion 2.5 mg/kg/h vs. sufentanil/midazolam:
| Measure | Change |
|---|
| Cardiac Index | -21% |
| SVR index | +38% |
| PCWP | +20% |
| mPAP | +14% |
| MAP | +13% |
This is the opposite of what a failing ventricle needs. The sympathomimetic effect failed to translate into improved cardiac output - instead, the marked SVR increase (afterload) further impaired an already compromised LV.
Why this happens in HFrEF (not in normal hearts):
Ketamine's indirect sympathomimetic effect requires catecholamine reserves to mobilize. In chronic HFrEF, the sympathetic system is already maximally activated, catecholamine stores are depleted, and the sympathetic nerve terminals are down-regulated. When there is no reserve to recruit, ketamine's direct myocardial depression goes unopposed, and the concurrent SVR elevation from residual sympathomimetic activity imposes further afterload on the failing LV.
Propofol-Ketamine Combination in Low EF (PMC4493732):
A study comparing propofol-ketamine vs. etomidate-midazolam for induction in CABG patients with LV dysfunction found both combinations produced comparable, acceptable hemodynamics. The ketamine component partially offset propofol's vasodilation and cardiac depression - but this was a brief induction dose scenario, not prolonged infusion. SVR was significantly different between groups only at 5 minutes post-intubation (p=0.009), with CI differences at 1-3 minutes. Both were deemed safe for induction.
Comparative Summary Table
| Feature | Sevoflurane | Propofol infusion | Ketamine infusion |
|---|
| Evidence level in HFrEF | Extrapolation from cardiac surgery RCTs/meta-analyses | Pharmacodynamic data + PRIS reports | Direct RCT (Christ 1997) in EF ~30% |
| Net effect on CO in HFrEF | Mild decrease; compensated by ↓SVR | Moderate decrease; blunted baroreflex | Significant decrease (-21%) |
| Effect on afterload (SVR) | Decreases (favorable) | Decreases (favorable) | Increases +38% (very unfavorable) |
| Effect on filling pressures | Decreases (↓ congestion) | Decreases (↓ congestion) | Increases PCWP +20% (worsens congestion) |
| Cardioprotection | Yes - anesthetic preconditioning (mechanistically strong; clinical translation variable) | Modest antioxidant effect; attenuates RIPC | None specific; raises MVO₂ |
| PRIS risk | None | Yes (prolonged high-dose) | None |
| Induction bolus safety | Requires slow titration | Requires reduced dose, vasopressors ready | Acceptable in extremis (brief dose only) |
| Maintenance infusion safety in HFrEF | Generally preferred; titrate to low MAC | Feasible with hemodynamic support | Avoid |
| Pulmonary hypertension concern | Not worsened | Not worsened | Worsened (mPAP +14%) |
| Overall rank in HFrEF | Most favorable | Intermediate (better than ketamine; needs careful titration) | Least favorable for infusion |
Preferred Anesthetic Approach in HFrEF (Evidence-Integrated)
Induction: Etomidate (minimal hemodynamic effect) remains the safest induction agent for HFrEF. If unavailable, a propofol-ketamine combination at reduced doses is acceptable. Ketamine alone is not preferred.
Maintenance:
- Sevoflurane (low MAC, 0.5-1.0): Best cardiovascular profile for HFrEF - modest afterload reduction, cardioprotective properties, titratable
- Propofol infusion (low rates, with hemodynamic monitoring): Acceptable, but no cardioprotective advantage; avoid high-dose prolonged infusion
- Ketamine infusion: Avoid based on direct RCT evidence of cardiac index reduction and increased filling pressures in the HFrEF population
Optimal strategy (from evidence): Low-dose sevoflurane with adjunctive opioid, invasive arterial monitoring, possibly transesophageal echocardiography guidance for real-time CO assessment. Vasopressor support (norepinephrine preferred over vasopressin for afterload-sensitive HFrEF; dobutamine if inotropy needed) should be immediately available.
Key References:
- Christ et al. 1997 (PMID: 9209606) - Ketamine infusion RCT in HF (EF 30%)
- Jiao et al. 2019 (PMID: 31661512) - Volatile vs. TIVA meta-analysis in CABG
- Guinot et al. 2020 (PMID: 33327246) - Sevoflurane did not reduce myocardial injury vs. propofol
- Zarbock et al. 2025 (PMID: 41093687) - Propofol attenuates RIPC; sevoflurane preserves conditioning
- Miller's Anesthesia 10e, p.2479-2482 - Propofol cardiovascular pharmacology
- Barash Clinical Anesthesia 9e, p.1428-1429 - Volatile cardioprotection