A practical method for reducing the interference due to lipaemia in coagulation tests

Abstract Introduction Plasma samples with gross lipaemia present a challenge for coagulation laboratories using optical analysers. High‐speed centrifugation may be used to remove excess lipids but it has not established whether this affects haemostasis tests. The aims were to determine whether the removal of lipid by centrifugation affects PT, APTT, fibrinogen, D‐dimer and von Willebrand factor activity measurements. Methods Twenty‐six lipaemic samples (median [range]): triglyceride 4.6 mmol/L [0.5‐17.0]; cholesterol: 4.06 mmol/L [2.20‐9.41] and 20 plasmas spiked with Intralipid 20 or lipid isolated from patient plasmas (median triglyceride of 11.95 mmol/L [5.0‐17.0] and cholesterol 4.33 [3.22‐7.06]), were tested before and after the removal of the lipid layer by centrifugation (10000 g for 10 minutes). Tests were performed using the CS‐5100 (Sysmex) coagulation analyser. Results Thirteen, 9, 3 and 1 of the lipaemic or spiked samples failed to give PT, APTT, fibrinogen and D‐dimer results, respectively. Centrifugation significantly reduced triglyceride (median 2.7, [0‐6.1 mmol/L]) and cholesterol (median 0.52 [0‐3.5]), allowing clot detection in all tests. There were no statistically significant differences in fibrinogen, D‐dimer or VWF levels in samples before and after lipid removal. A small but clinically insignificant change in PT and APTT was observed after lipid removal. Conclusion High‐speed centrifugation reduces lipaemia sufficiently to allow testing on an optical coagulation analyser without introducing clinically significant differences PT, APTT, fibrinogen, D‐dimer or VWF activity values.


| INTRODUC TI ON
Lipaemic plasma samples are commonly encountered in hospital diagnostic laboratories. Excessive lipaemia can increase turbidity in plasma samples due to an excess of large lipoprotein particles, especially chylomicrons. 1 Consequently, plasma samples with lipaemia present a challenge for clinical laboratories using optical analysers since light scatter and absorption may be increased resulting in reduced light transmission. 2 This may result in instrument error flags, requiring visual inspection of coagulation curves or, in the presence of gross lipaemia, instrument vote-outs (ie no result is generated).
Lipaemia may be the result of postprandial blood lipid changes or following lipid infusions in critically ill patients. In these cases, a repeat sample taken from a fasting patient or prior to lipid infusion may eliminate the problem, 3 but this leads to delay in obtaining results, which could delay diagnosis or treatment. However, if the lipaemia is due to a medical/metabolic condition (eg familial hypertriglyceridaemia, diabetes mellitus, renal and liver disease), repeat testing will not eliminate the problem. In these cases, the only alternative is retesting with a mechanical coagulometer or manual tube tilt-tube method in a water bath. However, many clinical laboratories do not have the facilities to do this.
Several methods have been proposed to reduce optical interference due to lipaemia, including ultracentrifugation (>60 000 g); high-speed centrifugation (8000-20 000 g); and lipid extraction using organic solvents or lipid-clearing agents. The Clinical and Laboratory Standards Institute (CLSI) 4 states that "ultracentrifugation has been suggested in some circumstances; however, although there is anecdotal evidence to suggest that this is widely used in haemostasis laboratories, there are no published studies." Ultracentrifugation is unavailable to most clinical laboratories, and furthermore, it may result in the loss of high molecular weight analytes such as fibrinogen, the FVIII/von Willebrand factor complex and D-dimer. 2  We propose a simple high-speed centrifugation method to reduce the level of lipaemia to a level which allows testing on an optical analyser without substantially altering the test results in a range of haemostasis parameters.

| Samples
Blood samples were collected into 0.109 M sodium citrate  For the statistical analysis (Table 1), the results of prespin lipaemic samples and prespiked plasmas were combined into a single group and compared with the results of postlipid depletion plasmas. Results for lipaemic samples which did not produce results before centrifugation were excluded from the analysis for the test that was affected. No significant differences in fibrinogen, D-dimer, VWF Ac or VWF:RCo were observed. PT and APTT results were slightly shorter but, although the differences were statistically significant, they were too small to be of clinical significance (Table 1). Excellent correlation was observed for all tests ( Figure 2).

| D ISCUSS I ON
Excessive lipaemia can present a challenge to diagnostic haemostasis laboratories using photo-optical coagulometers due to excessive turbidity. Current options available to the laboratory are retesting with a mechanical clot detection method or the removal of excess  is not available to most diagnostic laboratories. The only study to use an organic solvent (n-hexane) for haemostasis samples found significant changes in both PT and APTT. 6 We have investigated the use of high-speed centrifugation to produce results before centrifugation unspun. No clinically significant differences in the 6 parameters tested were observed after lipid depletion. It was notable that there were no statistically significant differences in high molecular weight analytes (fibrinogen, D-dimer, VWF Ac and VWF:RCo) as the result of high-speed centrifugation.
Correlation between haemostasis test results before and after lipid reduction was excellent in all six parameters. Although the reduction in triglyceride and cholesterol was modest, the reduction in turbidity, as assessed by visual inspection and elimination of instrument vote-outs, was manifested. This would suggest that the largest lipoprotein particles were preferentially removed by high-speed centrifugation as previously reported. 5 Our data show that if a result is obtained by the coagulometer, then that result does not change after reduction of lipid; thus, only samples that do not yield a result require high-speed centrifugation. The limitations of our study were that only one haemostasis analyser with a single set of reagents was studied. The technique described will therefore require validation for other analyser types and reagents. We believe that this simple method for reducing optical interference due to lipaemia may be suitable for most diagnostic laboratories.

CO N FLI C T O F I NTE R E S T
The Haemostasis Research Unit has received funding in the form of unrestricted educational grants from Sysmex UK. However, Sysmex UK had no part in the design of the study or preparation of this manuscript.