Checking potassium in haemolysed samples before suppressing the result can be an early warning of severe hypokalemia

The Working Group Preanalytical Phase of the European Federation of Clinical Chemistry and Laboratory Medicine advises suppressing haemolysis-sensitive tests when haemolysis is clinically significant, as improper specimen handling can rupture red blood cells, increasing potassium levels. Thus, a correctly repeated blood sample should show lower potassium levels than a haemolysed one. This study aimed to determine the prevalence of haemolysed samples with potassium levels below the reference range and whether this predicts hypokalemia in repeated collections. From March 2022 to March 2024, 396,640 non-haemolysed samples (H-index < 2 UA) were analyzed. Samples with an H-index ≥ 2 AU were classified as haemolysed, and potassium values were suppressed. Warnings were issued for haemolysed samples with potassium below the reference range (3.4-4.5 mmol/L), advising new sample collection. Twenty-five (0.01%) repeat samples were taken within 24 hours of a previously haemolysed sample with low potassium. Severe and moderate hypokalemia were more common in these repeats, with severe hypokalemia (≤2.5 mmol/L) found in 48% of repeat tests, compared to 0.3% in all samples. Though rare, a decrease in potassium in haemolysed samples often precedes hypokalemia diagnosis. Implementing a simple and cost-effective LIS algorithm to alert clinicians could potentially reduce diagnosis and treatment time.