Valve surgery in failing hearts: survival data reshape a high-risk decision

Patients with severe aortic valve disease and a failing left ventricle have long been considered too fragile for open-heart surgery, but a new retrospective cohort study from a Malaysian cardiac centre suggests the operation may deliver better medium-term survival than many clinicians expect. Researchers at Institut Jantung Negara reviewed 84 adults whose left ventricular ejection fraction was 35 percent or lower and who underwent surgical aortic valve replacement between 2015 and 2024. Fifty-three had severe aortic stenosis, a narrowing of the valve that obstructs blood leaving the heart, and 31 had severe aortic regurgitation, in which the leaky valve lets blood flow backward and volume-loads the ventricle. The findings, published in Cardiovascular Ultrasound, combine operative mortality figures with detailed echocardiographic and myocardial strain measurements taken before surgery and again six months afterward.

The numbers challenge the pessimism embedded in older literature. Among patients with severe aortic stenosis, in-hospital mortality was 9.4 percent, meaning five of 53 patients died before discharge. That figure rose to 13.2 percent at one year and 43.4 percent at a median follow-up of 55 months, or 4.6 years. In the regurgitation group, no patient died in hospital, one-year mortality was 3.2 percent, and mortality reached 32.3 percent at 55 months. Kaplan-Meier analysis put eight-year survival at 38.7 percent for the stenosis group and 43.1 percent for the regurgitation group. Nearly five years after surgery, 30 of 53 stenosis patients and 21 of 31 regurgitation patients were still alive.

Those outcomes stand in sharp contrast to what happens when such patients are managed without surgery. The authors cite earlier work showing that patients with severe aortic stenosis and reduced ejection fraction who did not undergo valve replacement had one-year mortality of 59 percent and five-year mortality of 85 percent, compared with 13.2 percent and 43.4 percent in this operated cohort. For severe regurgitation, conservative management carried one-year mortality of 35 percent versus 3.2 percent after surgery, and five-year mortality of 63 percent versus 32.3 percent. One cited report found a mean survival of only six months among regurgitation patients who never had the operation. The researchers caution that the comparison is observational and that their retrospective design cannot prove causation, but the magnitude of the gap is striking.

The study also delivered a sobering message about the measurements cardiologists routinely rely on. Despite comprehensive strain imaging, neither ejection fraction nor global longitudinal strain predicted mortality in either group. Global longitudinal strain, which tracks the shortening of heart muscle fibers during each beat, was markedly impaired before surgery, with median values in the single digits: minus 6.80 percent in the stenosis group and minus 8.80 percent in the regurgitation group. Left atrial reservoir strain, a marker of left atrial filling pressure, and right ventricular free-wall strain were also substantially reduced in both groups. Yet none of these strain parameters carried prognostic weight, and the authors suggest the narrow range of severely reduced ejection fractions, all below 35 percent, left the measure with no incremental predictive value within this already sick subgroup. They also acknowledge that the small sample size may have limited statistical power to detect such associations.

What did predict death was revealing. In the stenosis group, higher preoperative creatinine was associated with a 1 percent increase in all-cause mortality for each unit rise, a hazard ratio of 1.01 with a p-value of 0.022, pointing to the importance of kidney function and volume status before surgery. Conversely, a higher preoperative peak velocity across the narrowed valve was independently associated with a 48 percent reduction in mortality, with a hazard ratio of 0.52 and a p-value of 0.020. Peak velocity serves as a surrogate for forward flow, and the authors interpret higher values as a possible sign of preserved flow reserve and contractility. In the regurgitation group, only the length of the postoperative hospital stay predicted mortality, with each additional day raising the hazard by 13 percent, a hazard ratio of 1.13 and a p-value of 0.017, likely reflecting surgical complications or underlying frailty.

Six months after surgery, the two conditions recovered along distinctly different trajectories. Patients with stenosis showed considerable improvement: stroke volume index rose from 25.96 to 32.75 milliliters per square meter of body surface area, ejection fraction improved from 27.73 to 36.23 percent, and flow rate increased from 150.15 to 237.55 milliliters per second, all statistically significant. The authors report this as the first demonstration of flow-rate normalization after valve replacement, extending a metric traditionally used to grade stenosis severity into the postoperative recovery literature. Global longitudinal strain and left atrial reservoir strain improved numerically but did not reach statistical significance. Right ventricular free-wall strain, however, worsened significantly, from minus 13.61 to minus 9.71 percent, a change the authors attribute to the adverse impact of pericardial incision on right ventricular function during cardiac surgery.

Regurgitation patients followed a different path. Their ejection fraction remained essentially stable, slipping from 29.72 to 28.62 percent without statistical significance, while stroke volume index fell sharply from 65.97 to 42.66 milliliters per square meter and flow rate declined from 409.00 to 336.32 milliliters per second, both significant. The authors read this as normalization of volume overload rather than deterioration: the ventricle had been pumping an excessive regurgitant volume before surgery, and closing the leak removed that extra load. Strain parameters in this group showed no significant change. The researchers hypothesize that ejection fraction recovery after valve replacement for regurgitation may simply take longer than after surgery for stenosis, because the ventricle must adapt to the sudden removal of volume overload rather than the relief of pressure overload.

The etiological picture also diverged from earlier reports. Degenerative bicuspid valve disease caused 60.4 percent of severe stenosis cases, followed by degenerative tricuspid valve disease at 32.1 percent and rheumatic disease at 3.8 percent. Among regurgitation patients, aortic root dilatation led at 29.0 percent, prolapsed fused bicuspid leaflets accounted for 22.6 percent, and right coronary cusp prolapse and rheumatic disease each contributed 9.7 percent. The authors note that the most comprehensive prior study of regurgitation mechanisms, by Chaliki and colleagues, identified degenerative disease, bicuspid valve, and aortic root dilatation as the leading causes, a different pattern that hints at regional or temporal variation in disease and warrants further investigation.

Several limitations frame these findings. The retrospective design restricts causal inference, the study was not a head-to-head comparison of the two valve lesions, and the sample of 84 patients is small, reflecting how rarely such severely impaired patients are offered surgery at all. Echocardiographic follow-up stopped at six months, potentially missing slower remodeling that could emerge later. The stenosis and regurgitation groups also differed substantially in age, with a median of 63 years versus 44 years, which may partially explain the worse mortality observed in the stenosis cohort. The authors call for larger, prospective, multicenter studies with extended follow-up to confirm the absence of prognostic value in strain parameters and to clarify whether delayed recovery patterns exist.

Even with those caveats, the study carries a clear clinical signal. Operative mortality of 9.4 percent in stenosis and zero in regurgitation compares favorably with historical rates of 10 to 25 percent and 10 to 15 percent respectively for such patients, and medium-term survival substantially exceeds what conservative management has delivered. The authors emphasize that a 4.6-year mortality of 43.4 percent in stenosis and 32.3 percent in regurgitation remains high and should not be viewed as a panacea. Yet for a population long considered too high-risk to operate on, the data suggest that carefully selected patients with severe aortic valve disease and ejection fraction at or below 35 percent may gain meaningful survival from timely surgical intervention, with kidney function, forward flow, and recovery complexity guiding prognosis better than the ejection fraction itself.

Subject of Research: Outcomes of surgical aortic valve replacement in patients with severe aortic stenosis or regurgitation and left ventricular ejection fraction of 35 percent or lower

Article Title: Surgical aortic valve replacement in severe left ventricular dysfunction: mortality, echocardiographic, and strain outcomes

Article References: Aslannif, R., Soh, S. L., Tey, Y. S., Norhaliza, A. H., Jassie, T., Rowina, L. M., Chong, K. S., Wong, K. B., Jeswant, D., Alwi, M. Y., & Azmee, M. G. (2026). Surgical aortic valve replacement in severe left ventricular dysfunction: mortality, echocardiographic, and strain outcomes. Cardiovascular Ultrasound, 24(1), Article 10. https://doi.org/10.1186/s12947-026-00371-9

Image Credits: AI Generated

DOI: 10.1186/s12947-026-00371-9

Keywords: surgical aortic valve replacement, aortic stenosis, aortic regurgitation, left ventricular dysfunction, ejection fraction, global longitudinal strain, echocardiography, mortality, myocardial strain, cardiac surgery, survival outcomes, Surgical