Pitfalls in the Laboratory Assessment of Serum Prealbumin Levels in Hemato-Oncological Patients

Background: To highlight specific aspects of serum prealbumin measurements in hemato-oncological patients. Methods: Retrospective analysis of 4911 serum prealbumin measurements with special attention to values from hemato-oncological Intensive Care Units (ICU) patients. Results: Prealbumin serum levels in hemato-oncological ICU patients (n = 530) were significantly higher when compared to other ICU cohorts (p <0.0001). Their prealbuminemia did not correlate with serum albumin (p = 0.104) and was not affected by serum cholesterol or triglycerides (p = 0.076 and p = 0.430, respectively). Surprisingly, serum prealbumin has shown a positive correlation with cyclosporine in whole blood levels (r = 0.269, p = 0.010). Conclusions: A supposed explanation for our findings probably lies in a combination of several factors, including interference with the analysis itself and contamination or possible interference with the medication, e.g., cyclosporine. The prealbumin values do not reflect the actual nutritional state and cannot be regarded as a useful marker of malnutrition in these patients.