No Benefit to Conventional Ultrafiltration During Cardiac Surgery

Using conventional ultrafiltration during cardiac surgery did not appear to reduce the incidence of postoperative pulmonary complications, according to a retrospective analysis.

After adjusting for confounding variables, the incidence of the composite primary outcome -- reintubation, prolonged ventilation, pneumonia, or pleural effusion -- was 12.1% in the conventional ultrafiltration group versus 9.9% in the group who did not receive conventional ultrafiltration (estimated odds ratio 1.25, 95% CI 1.02-1.53, P=0.03), reported Andrej Alfirevic, MD, of the Cleveland Clinic in Ohio, and co-authors.

When giving equal weight to each component, the mean relative effect odds ratio for the pulmonary outcomes was 1.43 (95% CI 1.09-1.88, P=0.01) for patients receiving versus not receiving conventional ultrafiltration, they noted in the Annals of Thoracic Surgery.

"Conventional ultrafiltration was introduced to mitigate bypass-related effect of increased total body water and interstitial edema on the development of intraoperative anemia (thus decreasing allogeneic blood transfusion) and inflammation," Alfirevic and team wrote. "Consequently, previous work demonstrated improvement in pulmonary function related to total fluid and inflammatory cytokine removal. However it was predominantly the utilization of modified- and/or zero-balanced ultrafiltration, and not conventional ultrafiltration, that showed substantial benefit."

In a secondary analysis, the researchers found that use of conventional ultrafiltration was not associated with improved early pulmonary function.

The median arterial partial pressure to fractional inspired oxygen concentration ratio was 373 in the conventional ultrafiltration group versus 368 in the control group, with an estimated adjusted difference in medians of 5 (95% CI -5.9 to 16, P=0.37).

In addition, while the authors noted that data for postoperative and intraoperative blood transfusions and vasoactive-inotropic scores were missing, they said the estimated odds ratio of intraoperative red blood cell transfusion was 1.38 (95% CI 1.19-1.60, P<0.0001), and 1.30 (95% CI 1.14-1.49, P=0.0001) for postoperative red blood cell transfusion.

The estimated difference in medians of vasoactive-inotropic score between patients receiving and not receiving conventional ultrafiltration was 1.7 (95% CI 1.0-2.4, P<0.0001).

"Conventional ultrafiltration is used to preserve end-organ oxygen delivery and modulate the systemic inflammatory response," Alfirevic and colleagues wrote. "Ultrafiltration, as such, is considered standard of care in the pediatric population due to attenuation of the inflammatory pulmonary injury, lung edema and impaired pulmonary function, albeit using in-series, post-bypass, modified ultrafiltration. Conventional ultrafiltration, however, is less frequently used during adult cardiac surgery because the evidence for its effect on pulmonary complications is scarce."

There is still much to understand about ultrafiltration and its effects on outcomes, they noted.

"Previous work demonstrates lack of significant difference in lung injury regardless of the timing of ultrafiltration or the volume of ultrafiltrate. Consequently, the effect of conventional ultrafiltration exposure on pulmonary outcomes in the adult population remains unclear," they wrote. "The lack of standardized ultrafiltration technique is obvious and as such prevents comparison between the studies and ultimately leads to inconsistencies in patient care."

For this study, Alfirevic and team included 8,026 patients who underwent cardiac surgery with cardiopulmonary bypass at the Cleveland Clinic main campus from February 2017 through December 2020. Of these patients, 13% received conventional ultrafiltration. Mean age was 63, and 37% were women.

The authors noted that, due to the retrospective nature of the study, they could only determine association, and not causation. In addition, the lack of standardized use of conventional ultrafiltration may have introduced selection bias.

Comment