Pre-analytical quality control in hemostasis laboratories: visual evaluation of hemolysis index alone may cause unnecessary sample rejection

Abstract Background Visual inspection is the most widespread method for evaluating sample hemolysis in hemostasis laboratories. The hemolysis index (HI) was determined visually (visual index, VI) and measured on an ACL TOP 750 (IL Werfen) system with a hemolysis-icterus-lipemia index (HIL) module. T...

Full description

Saved in:
Bibliographic Details
Published in:Journal of laboratory medicine 2019-04, Vol.43 (2), p.67-76
Main Authors: Storti, Simona, Battipaglia, Elena, Parri, Maria Serena, Ripoli, Andrea, Lombardi, Stefania, Andreani, Giovanna
Format: Article
Language:eng ; ger
Subjects:
Automation
Citric acid
Correlation analysis
Diagnostic systems
Hemoglobin
Hemolysis
hemolysis index
Hemostasis
hemostasis laboratory
Hemostatics
Inspection
Jaundice
Laboratories
Mathematical analysis
Measurement methods
Multivariate analysis
Organic chemistry
pre-analytical quality control
Quality control
Quantitation
Regression analysis
Sample rejection
Sensitivity analysis
Sodium
Sodium citrate
Spectrophotometry
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract Background Visual inspection is the most widespread method for evaluating sample hemolysis in hemostasis laboratories. The hemolysis index (HI) was determined visually (visual index, VI) and measured on an ACL TOP 750 (IL Werfen) system with a hemolysis-icterus-lipemia index (HIL) module. These values were compared with those measured on clinical chemistry systems Unicel DXC600 and AU680 and with quantitation of free-hemoglobin (Hb) performed by a spectrophotometric measurement method (SMM). Methods The HI was measured in 356 sodium citrate plasma samples, 306 of which were visibly hemolyzed to varying degrees and 50 were not hemolyzed. The analytical performance of each method was evaluated. Results Linear regression analysis, calculated between SMM and the other systems in the study, returned coefficients of determination r 2  = 0.853 (AU680), r 2  = 0.893 (DXC600) and r 2  = 0.917 (ACL TOP 750). An r 2  = 0.648 was obtained for linear regression analysis between VI and ACL TOP 750. In addition, ACL TOP 750 showed an excellent correlation in multivariate analysis (r 2  = 0.958), showing good sensitivity (0.939) and specificity (0.934) and a diagnostic accuracy of 94%. By comparison, DXC600 and AU680 showed coefficients of determination of 0.945 and 0.923, respectively. A cut-off was set at 0.15 g/L free-Hb, as determined by the automated method, such that any hemostasis samples measuring above this threshold should not be analyzed. Based on this criterion, samples were classified as accepted or rejected, and the number of samples discarded during VI or ACL TOP 750 measurements was compared. Conclusions This study confirmed that hemostasis laboratories should consider introducing an objective, automated and standardized method to check samples for hemolysis. By relying solely on visual inspection, up to 50% of samples could be unnecessarily rejected. The ACL TOP 750 system demonstrated a satisfactory analytical performance, giving results comparable to those of the reference method.
ISSN:0342-3026
2567-9430
1439-0477
2567-9449
DOI:10.1515/labmed-2018-0122