Clarifying clinical impact and methodological considerations in the comparison of OPAL and ALBUNIQUE systems

To editor,

I read with great interest the recent article comparing the extracorporeal albumin dialysis (ECAD) systems OPAL and ALBUNIQUE in critically ill patients with liver dysfunction. ¹ Several concerns have been raised.

The retrospective and non-randomized nature of treatment allocation introduces a substantial risk of selection bias. As acknowledged by the authors, the choice between OPAL and ALBUNIQUE was based on device availability and clinicians’ discretion. ¹ This approach may have resulted in unmeasured confounding, particularly if one system was preferentially used in patients with differing hemodynamic stability, severity of liver dysfunction, or expected prognosis. Additional analyses adjusting for time-varying severity, such as SOFA trends or vasopressor requirements, or sensitivity analyses restricted to first treatment cycles might help mitigate such a concern.

The interpretation of “no significant difference” between devices requires caution. The study included only 25 patients and 90 treatment cycles, with a relatively small number of ALBUNIQUE sessions (n = 32). ¹ Given this limited sample size, the study may be underpowered to detect clinically meaningful differences. ² Equivalence or non-inferiority cannot be inferred from non-significant findings alone, and reporting of effect sizes with confidence intervals should be emphasized to better contextualize the results.

The reliance on surrogate biochemical endpoints, such as bilirubin and ammonia reduction, may not fully capture the clinical impact of ECAD therapy. While reductions in these parameters were demonstrated, key patient-centered outcomes—including survival, organ recovery, duration of intensive care unit stay, and improvement in hepatic encephalopathy—were not systematically evaluated. ¹ Notably, the reported intensive care unit mortality remained high (68%), raising the question of whether biochemical improvements translate into significant clinical benefit. ³

Important mechanistic parameters were not assessed. As the authors note, data on bile acid removal and albumin binding capacity were not stated. ¹ These variables are central to the theoretical advantage of albumin dialysis systems and may represent key differentiators between devices. ⁴ Incorporating these measurements in future studies would provide deeper insight into device-specific efficacy.

The analytical approach using generalized estimating equations appropriately accounts for repeated measures. However, cycle-based analysis may obscure patient-level effects. Given that multiple treatment cycles were clustered within individuals, complementary patient-level analyses or hierarchical modeling could help clarify whether observed effects are consistent across patients or driven by a subset with repeated treatments. ⁵