| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Technical Briefs |
1 Department of Chemical Pathology, Queensland Health Pathology Service, Princess Alexandra Hospital, Woolloongabba, Queensland Australia
2 Department of Chemical Pathology, Sullivan and Nicolaides Pathology, Taringa, Queensland, Australia
aaddress correspondence to this author at: Department of Chemical Pathology, Queensland Health Pathology Service (QHPS), Princess Alexandra Hospital, Ipswich Road, Woolloongabba, Queensland 4102, Australia; fax 61-7-3240-7070, e-mail Goce_Dimeski{at}health.qld.gov.au
The new Roche IFCC-recommended method for lactate dehydrogenase (LD) has been reported (1)(2) to produce duplicate errors with Becton Dickinson (BD) lithium-heparin tubes with plasma separators. Separate studies (3) of Sarstedt tubes showed no such errors. To overcome and minimize duplicate errors, Roche has modified the method to incorporate a predilution step. This modified method can be used only on dilutor systems (Modular P and Hitachi 911, 912, and 917), but not on nondilutor systems (Modular D and Hitachi 747 and 902; Roche). We tested the modified method for duplicate errors and tested lithium-heparin tubes (with plasma separator) from another major supplier for duplicate errors.
Blood samples were collected in three different tubes: Greiner Vacuette lithium-heparin tubes with plasma separator (prod. no. 455083); Greiner Vacuette serum tubes with plasma separator (prod. no. 455078); and BD Vacutainer PST II lithium-heparin tubes with plasma separator (prod. no. 367377). Three IFCC LD methods were set up on our Modular DP system: the Modular D, the Modular P IFCC method, and the Modular P IFCC modified method with predilution. The two LD methods on the Modular P system were set up to use the same reagent bottle set. The three tubes for each sample were collected, mixed without delay by inversion, centrifuged, and analyzed at the same time. After centrifugation at 3000g for 12 min at 4 °C, the samples were immediately analyzed in duplicate first on the Modular D system, then on the Modular P by the two methods. The study was conducted over 4 days with 51 samples being analyzed in groups of 5 or 10. The LD range in the samples was 145–564 U/L.
Using the formula suggested by Bakker et al. (1), in which the 95% confidence limit = 0.028 times the mean of the duplicate, we calculated the frequency of duplicate errors for each tube/method. At the same time, we performed statistical analyses of the duplicate results, using the Bland–Altman procedure (4). The observed absolute difference range between the duplicate readings was also included. The results are shown in Table 1
.
|
The data show that Greiner lithium-heparin tubes produce a frequency of duplicate errors similar to the frequency of errors obtained with the BD tubes. From the data obtained, use of the modified Modular P IFCC method substantially reduced but did not totally eliminate duplicate errors.
For many years, we have observed on an intermittent basis duplicate errors when using lithium-heparin gel tubes (5). These have been most prominent with analytes such as LD, phosphate, sodium, chloride, and potassium on the Hitachi series analyzers. These errors are attributable to cell aggregates, or microclots, which can be throughout the sample but are typically seen floating at the top of the meniscus and almost exclusively in lithium-heparin plasma. In the most recent study we conducted with Greiner (prod. no. 455083) and BD (prod. no. 367377) lithium-heparin tubes with a new gel formulation, we found that 8% and 4%, respectively, of tubes had microclots present on top of the meniscus that were visible to the naked eye (Fig. 1
). We wish to highlight that both lithium-heparin tubes have gel formulations that are not microclot free. The microclot problem is dramatically reduced in serum samples.
|
In the Hitachi analyzers, the analyte sample volume ranges from 2 to 32 µL. To aspirate the sample volume, the probes descend only 2–3 mm into the sample below the meniscus. The presence of cell aggregates (fibrin as well as erythrocytes, leukocytes, or platelets) or microclots can cause sampling problems. This can lead to short sampling if the probe picks up the designated sample volume with a microclot as part of the total sample volume. Alternatively, if the microclot disintegrates, the cells may rupture as a result of reagent (pH) changes, releasing the cellular contents (e.g., LD, phosphorus, and aspartate aminotransferase). In contrast, if the sample probe picks up a microclot (plus or minus cell aggregates), and the microclot hangs on the outside of the sample probe, then an additional sample volume may be aspirated and dispensed. Additionally, intact microclots in the reaction cuvette may produce optical interferences as the microclots floats in front of the photometric light path. The potential of microclots to alter the accuracy of assays such as LD using plasma samples is ever present. Vigilant scans by operators of primary lithium-heparin plasma samples from the top is essential.
The Modular P IFCC modified method with a predilution step minimizes duplicate errors by diluting the sample in saline and then resampling. This reduces the risk of reaspirating the same microclots that may have been aspirated in the primary sampling. However, the potential for inaccuracies with the actual primary sampling as a result of microclots, as described above, are not eliminated even if the duplicate readings are within expected limits.
Finally, the Greiner and BD lithium-heparin tubes with plasma separators produced similar duplicate error frequencies. The use of serum samples is shown by the formula suggested by Bakker et al. (1) to substantially reduce duplicate errors, and the difference between the duplicate readings (Table 1
) would not be considered clinically significant. Additionally, the modified method with predilution will decrease the throughput of the analyzer, although minimally. Therefore, serum would be the sample of choice to avoid such duplicate errors and eradicate the need to implement the modified method.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
