International
Perfusion Association

Sevoflurane vs Propofol During CPB: Early Kidney Barrier Effects

Conceptual medical illustration of a kidney model between an anesthetic vaporizer and syringe pump, with bypass equipment behind

Sevoflurane vs propofol during CPB may produce different early effects on the kidney’s filtration barrier. In a randomized trial of 106 adults undergoing elective cardiac surgery, sevoflurane was associated with less urinary albumin leakage and smaller changes in several measures of glomerular barrier dysfunction. These biological differences did not establish a reduction in postoperative acute kidney injury (AKI).

Study Design and Population

Feng and colleagues conducted a prospective, single-center trial in China, enrolling patients during 2023. Fifty-three participants received sevoflurane-based anesthesia and 53 received propofol-based total intravenous anesthesia. Allocation was concealed; patients, postoperative intensive care physicians, laboratory personnel, and data analysts were blinded, although treating anesthesiologists could not be.

Most participants underwent valve surgery: 42 in the sevoflurane group and 45 in the propofol group. The remaining patients underwent coronary artery bypass grafting. Patients with preoperative serum creatinine above 133 μmol/L, previous cardiac surgery, planned aortic dissection surgery, or preoperative hemodynamic instability requiring continuous vasoactive support were excluded. All randomized participants received their assigned intervention and completed follow-up.

The primary endpoint was the perioperative trajectory of the urinary albumin-to-creatinine ratio (UACR). Blood and urine sampling continued through 24 hours after surgery, while postoperative AKI was monitored through day 7.

What the Kidney Measurements Mean

The glomerular filtration barrier normally restricts the passage of proteins into urine. UACR measures albumin leakage but does not identify which component of that barrier is affected. The investigators therefore combined UACR with circulating heparan sulfate and syndecan-1, markers related to glycocalyx shedding, and clearance-derived indices of charge and molecular-size selectivity.

Higher charge and size selectivity indices represented greater impairment in this study. These are exploratory functional measurements, rather than direct images of kidney injury or validated substitutes for clinical renal recovery. Circulating glycocalyx fragments also cannot identify the kidney as their sole source.

Sevoflurane vs Propofol During CPB: Main Findings

The longitudinal UACR pattern differed between groups (group-by-time interaction, P = 0.001). After aortic unclamping, estimated UACR was approximately 17–22% lower with sevoflurane than with propofol.

Sampling time Sevoflurane-to-propofol UACR ratio 95% confidence interval
10 minutes after aortic unclamping 0.785 0.714–0.863
1 hour after surgery 0.826 0.751–0.908
24 hours after surgery 0.779 0.708–0.856
Ratios compare model-estimated UACR between groups. Values below 1 favor lower albumin leakage with sevoflurane; all three Holm-adjusted P values were <0.001.

Heparan sulfate and the charge and size selectivity indices showed coordinated longitudinal differences favoring sevoflurane. The overall syndecan-1 interaction was not statistically significant (P = 0.078), even though some individual postoperative comparisons were lower with sevoflurane. The findings therefore should not be described as uniform protection across every glycocalyx marker.

Post hoc adjustments for selected baseline imbalances generally preserved the principal barrier findings. These sensitivity analyses strengthen the biological signal but do not establish its precise molecular mechanism.

No Proven Reduction in Acute Kidney Injury

AKI occurred in 13 of 53 sevoflurane recipients (24.5%) and 17 of 53 propofol recipients (32.1%; P = 0.518). The risk ratio was 0.76, with a 95% confidence interval of 0.41–1.41. That interval allows substantial uncertainty, including the possibility of benefit or harm.

Creatinine-based estimated glomerular filtration rate trajectories did not differ significantly. Mechanical ventilation duration, intensive care stay, hospital stay, and reported respiratory and cardiovascular event rates also showed no statistically significant between-group differences. The trial was powered for longitudinal UACR, rather than clinical renal outcomes; a nonsignificant AKI comparison does not prove equivalence.

Relevance for Perfusion Teams

During bypass, the sevoflurane group received the volatile anesthetic through a vaporizer and air-mixing system connected to the oxygenator, with concentration monitored at the outlet. Standardized management included nonpulsatile pump flow of 2.2–2.4 L/min/m², a CPB mean arterial pressure target of 50–70 mmHg, and mild-to-moderate systemic hypothermia.

The study illustrates why anesthetic exposure during bypass belongs in the interpretation of perioperative renal research. It did not randomize pump flow, pressure, oxygen delivery, or oxygenator type, and it cannot identify an optimal perfusion strategy. Its findings also do not establish that any oxygenator is suitable for volatile-anesthetic delivery; equipment compatibility and delivery procedures require their own assessment.

For clinical teams, the practical implication is to distinguish an early physiological signal from a demonstrated patient benefit. These results provide a rationale for further research, rather than sufficient evidence to change anesthetic selection solely to prevent AKI.

Limitations and Disclosures

This small, single-center study largely involved elective valve surgery and excluded patients with important baseline renal dysfunction or hemodynamic instability. Generalizability to emergency surgery, severe chronic kidney disease, and more complex aortic procedures is limited. Barrier sampling ended at 24 hours, so persistence of the differences remains unknown.

Both groups received propofol for early intensive care sedation, potentially reducing later separation. Clearance-derived indices cannot localize a structural lesion, and exploratory biomarker associations do not demonstrate causation. The authors report no conflicts of interest. Funding came from the Wenzhou City Science and Technology Bureau, which the paper states had no role in study conduct, analysis, or publication decisions.

Source

Feng L, Lin Y, Xi W, et al. Differential Effects of Sevoflurane and Propofol on Perioperative Glomerular Permselectivity During Cardiopulmonary Bypass: A Randomized Trial. Drug Design, Development and Therapy. 2026;20:651212. Published November 10, 2026. doi:10.2147/DDDT.S651212.

Full Text Free (PDF) Download Full Text PDF

Study Ranking

3
Randomized, prospectively registered single-center trial with concealed allocation, blinded outcome assessment and complete follow-up in 106 adults. Consistent early surrogate renal barrier findings support a mechanistic signal, but the small sample, predominantly valve-surgery population, exploratory clearance indices and lack of power for clinical AKI or renal recovery limit practice-changing conclusions.
Notifications · Choose topics or turn off