Clin Chem Lab Med 2015; aop

Janne Cadamuro*, Thomas Klaus Felder, Hannes Oberkofler, Cornelia Mrazek, Helmut Wiedemann and Elisabeth Haschke-Becher

Relevance of EDTA carryover during blood collection DOI 10.1515/cclm-2014-0944 Received September 24, 2014; accepted December 8, 2014

Abstract Background: The order of draw is regarded as a preanalytical issue to prevent carryover of additives during blood collection. Our objective was to prove the theory of ethylenediaminetetraacetic acid (EDTA) carryover for a closed vacuum system and the influence of EDTA on concentrations of selected biomarkers. Methods: To test the carryover of EDTA, a blood collection with tripotassium EDTA (K3EDTA) and subsequent non-additive tubes was simulated using distilled water as substitute for blood. EDTA concentrations were measured by tandem mass spectrometry. Then we added increasing concentrations of EDTA to heparinized blood and measured routine biomarkers, thereby simulating a carryover of EDTA whole blood and pure EDTA, respectively. Additionally, we tested for EDTA contamination and biomarker alteration in samples collected from 10 healthy volunteers by a syringe with subsequent transfer into sample tubes. Results: No EDTA contamination was detected in samples collected subsequent to a K3EDTA tube when adhering to guidelines of blood sampling. Magnesium, calcium, and potassium levels were altered by artificial K3EDTA wholeblood contamination as well as when adding 1 μL pure K3EDTA. Iron values were altered at EDTA concentrations of 4.4 mmol/L. All other parameters remained unaffected. A slight EDTA carryover was observed in syringe collection

*Corresponding author: Janne Cadamuro, MD, Universitätsinstitut für Medizinisch-Chemische Labordiagnostik, Gemeinnützige Salzburger Landeskliniken Betriebsges.m.b.H., Müllner Hauptstraße 48, 5020 Salzburg, Austria, Phone: +43-662-4482-57263, Fax: +43-0662-4482-885, E-mail: [email protected] Janne Cadamuro, Thomas Klaus Felder, Hannes Oberkofler, Cornelia Mrazek, Helmut Wiedemann and Elisabeth Haschke-Becher: Department of Laboratory Medicine, Paracelsus Medical University, Salzburg, Austria

and subsequent transfer into EDTA and heparin tubes, however, without any biomarker alteration. Conclusions: An EDTA carryover during blood collection using a closed vacuum system is highly unlikely. Even if carryover of EDTA whole blood occurs, an absolute volume larger than 10 μL would be necessary to alter test results. However, contamination of samples with preloaded pure K3EDTA solution by severe neglect of current recommendations in blood collection may significantly alter testing results. Keywords: blood sampling; order of draw; preanalytical phase; sample handling; specimen handling.

Introduction Accurate laboratory analyses are crucial for diagnosis, treatment, and prognosis of diseases in today’s healthcare system. In 64% of cases, laboratory results essentially contribute to diagnosis and clinical decision making [1]. Therefore, it is important that laboratory analyses are of high quality and errors during the preanalytical, analytical, and postanalytical phases are reduced to a minimum. When evaluating errors in these processes, the majority (68.2%) are to be found in the preanalytical phase [2]. The types of errors range from wrong sample identification to inappropriate sample containers to mistakes during sample collection. One of these errors is supposed to be the incorrect “order of draw”, based on the finding of Calam and Cooper [3] in 1982 who proposed that the sample collection, using evacuated tubes, should be performed in a certain order to prevent the carryover of additives. The supposed reason for these carryovers are additives adherent to the plunger of the evacuated tube holder [4]. Carryover of additives could potentially lead to falsely elevated or reduced laboratory parameters. Contamination of potassium ethylenediaminetetraacetic acid (K3EDTA) can lead to elevated levels of potassium as well as reduced levels of calcium, zinc, and magnesium [3, 5–7] and can even affect parameters like unsaturated ironbinding capacity, bicarbonate, aspartate transaminase,

Brought to you by | University of Michigan Authenticated Download Date | 5/20/15 1:36 AM

2      Cadamuro et al.: EDTA carryover during blood collection

alanine transaminase, lactate dehydrogenase, creatine kinase, alkaline phosphatase, and α-amylase [8]. Citrate contamination may cause similar effects [3, 5, 6]. Heparin contamination of citrated blood from a previously attached tube could potentially affect coagulation parameters [9]. Also, hematological parameters could be altered by fluoride or oxalate because these agents are disruptive to cell membranes [3, 10]. These findings found their way into the current World Health Organization (WHO) and Clinical and Laboratory Standards Institute (CLSI) recommendations and became part of preanalytical trainings all over the world [11, 12]. However, many of the studies in this field are based on few cases in which the phlebotomist often had difficulties in collecting the specimen [3]. Additionally, all the studies were performed in blood, with the potential bias that the effect on the analyte is due to causes other than additive contamination. Also, in many of these studies, it is not clear if blood sampling was performed according to current WHO and CLSI guidelines. For example, blood collecting into a syringe with subsequent transfer into different sample tubes could potentially lead to additive contamination [13]. In this study, we aimed to investigate the principle of EDTA carryover during venous blood draw when performed in a closed vacuum system according to current guidelines and the influence of EDTA on concentrations of selected biomarkers by simulating the specimen collection with water as well as simulation of EDTA contamination of human blood. Additionally, we aimed to prove the theory of Cornes et al. [14], who propose that EDTA contamination might occur with open blood collection systems by needle contamination when delivering collected blood from a syringe into sample tubes, thereby severely disregarding current recommendations on blood collection.

was collected into a non-additive tube (2 mL, Ref. 454088; Greiner BioOne). The second tube was analyzed by liquid chromatographytandem mass spectrometry (LC-MS/MS), as described below. To exclude a random error, we repeated the sample collection and the LC-MS/MS measurements 10 times. Additionally, we repeated our trial with the corresponding standard tube from Becton Dickinson (3 mL, Ref. 367652; Becton Dickinson Diagnostics), in which K3EDTA is applied as a liquid.

Mixing experiment Next, we performed an additional set of experiments, simulating an additive contamination by adding EDTA at different concentrations into heparinized whole-blood samples. These samples were collected from two healthy volunteers, using lithium-heparin tubes (Ref. 454049, GreinerBioOne). These volunteers, as well as the volunteers mentioned below, gave informed consent and the work was approved by our institution’s responsible committee. All laboratory parameters determined were within the reference range. The experiment was divided into three parts (Table 1). Part 1 was performed to determine the amount of EDTA needed to alter test results, even if these amounts are not realistic in a clinical setting. For this purpose, we calculated the amount of K3EDTA per microliter of whole blood in a properly filled 3-mL EDTA collection tube and simulated a carryover of 1 μL, 5 μL, 10 μL, 100 μL, and 1 mL of EDTA whole blood into heparinized blood. In part 2, we aimed to rule out any cause of influence not directly associated with EDTA by repeating part 1 of our experiment, replacing the K3EDTA from the BD tubes with disodium EDTA (Na2EDTA) (Sigma Aldrich). In part 3, we determined the effect of undiluted K3EDTA on the same laboratory parameters. To minimize alterations due to dilution effects during these mixing procedures, the volume of added EDTA was always kept at 24 μL/mL whole blood. In all samples, the following parameters were analyzed in duplicate: sodium, potassium, magnesium, calcium, iron, α-amylase,

Table 1 Experimental setup. Parts 1 and 2 Assumed carryover of  EDTA whole blood

Materials and methods Blood sampling simulation Blood sampling was simulated by collecting HPLC-grade distilled water (Sigma Aldrich, St. Louis, MO, USA) as substitute for blood. The water was drawn from a disposable 5-mL falcon tube (Ref. 352054; Becton Dickinson Diagnostics, Franklin Lakes, NJ, USA) using a 21-gauge needle (Ref. 450076; Vacuette Multiple Use Needle; Greiner BioOne International AG, Kremsmünster, Austria) in combination with a safety tube holder (Ref. 450230, Vacuette Quickshield; Greiner BioOne). A closed vacuum system was maintained over the entire simulation. As a first tube, a standard evacuated blood collection tube (3 mL, Ref. 454086; Greiner BioOne) was filled to the marking spot, yielding 4883 μmol/L K3EDTA. Thereafter, distilled water

1 μL 5 μL 10 μL 100 μL 1 mL

Calculated final EDTA concentrationa, μmol/L

         

4.4 22 44 440 4398

Amount of pure EDTAa  

Calculated final EDTA concentration, μmol/L

Part 3

1 μL 5 μL 10 μL

     

183 916 1833

a Parts 1 and 3: 7.5% K3EDTA from evacuated blood collection tubes (Becton Dickinson); part 2: Na2EDTA (Sigma Aldrich).

Brought to you by | University of Michigan Authenticated Download Date | 5/20/15 1:36 AM

Cadamuro et al.: EDTA carryover during blood collection      3

creatinine kinase, aspartate transaminase, and alanine transaminase. All analyses were performed on the Roche MODULAR-P analyzer (Roche Diagnostics, Rotkreuz, Switzerland). Results obtained from samples with added EDTA were compared with the corresponding results of the native sample. A randomly assigned difference cutoff of ±10% was used to define parameter alterations due to the added EDTA reagent.

transition for underivatized EDTA. The transition of MS/MS detection for the internal standard EGTA was m/z 381.1/160.1. Analyst software version 1.5.2 was used for data collection and analyses. The limit of detection (LOD) for the plasma method was 2 μmol/L EDTA and the cutoff value was 5 μmol/L EDTA, whereas the LOD for the simulated blood was determined to be 0.5 μmol/L EDTA by manually generating extracted ion chromatograms of the above-mentioned ion transitions.

Blood collection by syringe Statistical analyses The theory of EDTA carryover by needle contamination when delivering collected blood from a syringe into sample tubes was tested in 10 healthy volunteers. Blood was collected in the following order: (1) Li-heparin gel tube, (2) EDTA tube, and (3) Li-heparin gel tube. Subsequently, blood was drawn into a syringe (20 mL, Injekt; B-Braun, Melsungen, Germany) and transferred into (1) a Li-heparin tube, (2) an EDTA tube, and (3) into another Li-heparin tube using a 20-gauge needle (Vacuette Multiple Use Needle; Greiner BioOne). All tubes were decapped and filled to the marking spot within 1 min after blood collection. In every tube, the needle was intentionally dipped into the respective whole blood. In five of the participants, the described blood collection was performed using blood collection tubes from Greiner BioOne; in the remaining five participants, the respective BD tubes were used. In all heparin samples, chemistry parameters were analyzed as described above. An aliquot of each heparin plasma sample was collected and stored at –80 °C for batch analysis of EDTA by LC-MS/MS as described below.

Mass spectrometry After appropriate mixing, 5 μL of samples from the blood sampling simulation were injected directly into the LC-MS/MS system. Additionally, EDTA carryover was determined in human plasma samples by a semiquantitative LC-MS-MS method. Therefore, EDTA-free plasma was used to prepare calibrators and quality controls. 300-μL of a precipitation solution [80/20 acetonitrile/methanol (vol/vol), containing the internal standard EGTA (300 μmol/L EGTA)] were added to 100 μL of plasma/calibrator/QC. After vortexing and centrifugation (20,000 g, 10 min) 400 μL of mobile phase A were added to 100 μL of the clear supernatant. 5-μL of the solution were injected onto a monolithic silica Chromolith RP-18e 100-3 mm column (Merck KGaA). An Agilent 1200 Series HPLC system was used with 1 mmol/L ammonium formate/0.1% formic acid as mobile phase A and 1  mmol/L ammonium formate/0.1% formic acid in 95/5 (vol/vol) acetonitrile/ water as mobile phase B. Isocratic analysis was performed at a flow rate of 400 μL/min at 10% B for 3 min. For plasma samples, a gradient method was applied at a flow rate of 700 μL/min with 10% B for 0.7 min, rising to 100% B in 1.6 min, followed by a 0.5-min wash at 100% B and a 0.9-min re-equilibration step. LC-MS/MS analysis of underivatized EDTA was performed using an API4000 triple quadrupole mass spectrometer (AB SCIEX, Framingham, USA), equipped with a turbo IonSpray heated at 400 °C and operated in positive ion mode at an ESI spray voltage of 5000 V as published earlier [15, 16] with some modifications. Settings used were curtain gas 30, ion source gas 1 (GS1) 40, and ion source gas 2 (GS2) 40. The quantifier ion transition of MS/MS detection was m/z 293.1/160 (collision energy 21 V), whereas 293.1/132.1 (collision energy 31 V) served as qualifier

Statistical analyses were performed with SPSS using the Wilcoxon test to assess the significance between analyte values of samples.

Results Blood sampling simulation In our mass spectrometry assay, we were able to detect free EDTA in water samples with a LOD of 0.5 μmol/L. When measuring the samples, in which we simulated a sample collection with an EDTA tube and subsequent non-additive tube using distilled water, we could not detect any traces of EDTA, neither in Becton Dickinson nor in Greiner BioOne collection tubes, thus defining the concentration of EDTA that is potentially carried over during sample collection, using evacuated collection tubes, as < 0.5 μmol/L.

Mixing experiment In part 1, we found creatinine kinase, aspartate transaminase, alanine transaminase, and sodium to be completely unaffected by any EDTA concentration tested. Iron values were reduced by 18% at EDTA concentration of 4398 μmol/L. Magnesium and calcium levels were reduced by 13% and 48% at EDTA concentrations of 44 and 440 μmol/L, respectively, corresponding to a simulated carryover of 10 and 100 μL, respectively. A 100% decrease in magnesium and calcium values was reached at a simulated EDTA carryover of 1 mL or an EDTA concentrations of 4398 μmol/L. Measured potassium was elevated by 76% at simulated carryover of 100 μL K3EDTA-contaminated whole blood (Figure 1). The findings of part 2 of our mixing experiment showed similar results to those of part 1 for magnesium, calcium, iron, α-amylase, creatinine kinase, aspartate transaminase, and alanine transaminase. The only expected differences consisted in potassium being unaffected by Na2EDTA concentrations and elevated sodium

Brought to you by | University of Michigan Authenticated Download Date | 5/20/15 1:36 AM

4      Cadamuro et al.: EDTA carryover during blood collection

150

100

100

100

10 0 10 00

10

1

5 5

0

5

10 10 0 10 00 Sodium, %

100

50

10 10 0 10 00

μL carryover

0

5

1

0

10 10 0 10 00

μL carryover

μL carryover 150

200 150 100 50 0

10 10 0 10 00

5

1

0

0

800 700 600 500

0

Potassium, %

150

50

1

μL carryover

μL carryover

100

50

0

0

10 10 0 10 00

5

0

0

1

0

Iron, %

150

50

10 10 0 10 00

μL carryover

150

50

1

0

10 0 10 00

5

10

1

0

5

1

10 10 0 10 00

μL carryover

150

GPT, %

GOT, %

μL carryover

Magnesium, %

50

0

0

0

0

50

100

1

50

100

5

100

Creatine kinase, %

150 Calcium, %

Amylase, %

150

μL carryover

Figure 1 Changes of biomarker values after a simulated carryover of 1, 5, 10, 100, and 1000 μL of K3EDTA whole blood. Changes in parameter values, depicted in percent deviation. Dotted lines indicate an analytical variability ( ± 10%), error bars refer to the standard errors. AST/GOT, aspartate transaminase; ALT/GPT, alanine transaminase.

levels at simulated carryover of 1  mL Na2EDTA contaminated whole blood. In part 3, 1 μL of pure K3EDTA was sufficient to reduce calcium and magnesium concentrations by 20% and 32%, respectively. The same amount of K3EDTA, corresponding to concentrations of 183 μmol/L, elevated potassium values by 35%. All other parameters tested remained unaffected up to the addition of 10 μL of pure K3EDTA (Figure 2).

five samples; however, all of which under the cutoff level of 5 μmol/L. In the corresponding BD tubes, EDTA was found in all of the samples; in three of these, concentrations were slightly above the cutoff level. Measured biomarker values in samples collected prior to EDTA samples, compared with the corresponding sample, collected thereafter, showed no significant differences in any of the analytes.

Blood collection by syringe

Discussion

In all heparin samples collected subsequent to an EDTA tube, no traces of EDTA could be found when collection was performed into evacuated tubes according to guidelines. In samples collected into a syringe with subsequent transfer into collection tubes, EDTA could be detected in concentrations above the LOD of 2 μmol/L in some of the samples filled after the EDTA tube. In the respective tubes from Greiner BioOne, EDTA could be found in three of the

Preanalytical errors account for about 68% of all causes potentially leading to false laboratory results in today’s health care [2]. One of which is claimed to be the additive carryover during blood collection using evacuated collection tubes [4]. In theory, additives such as EDTA or heparin can potentially be carried over to the next tube in order. When collecting an EDTA blood sample, and subsequently a heparin sample, EDTA could be transferred

Brought to you by | University of Michigan Authenticated Download Date | 5/20/15 1:36 AM

Cadamuro et al.: EDTA carryover during blood collection      5

50

100

50

1 5 10 μL carryover

100

50

0

0 0

0

1 5 μL carryover

10

150

150

100

100

100

50

0

Iron, %

150

GPT, %

50

0 0

5 1 10 μL carryover

1 5 μL carryover

Potassium, %

100

50

1 5 μL carryover

0

5 1 10 μL carryover

0

5 1 μL carryover

10

50

10

500

150

0

0 0

150

400 Sodium, %

GOT, %

Creatine kinase, %

100

0

Magnesium, %

150

150 Calcium, %

Amylase, %

150

300 200

100

50

100 0

0

0 0

5 10 1 μL carryover

0

1 5 μL carryover

10

10

Figure 2 Changes of biomarker values after addition of 1, 5, and 10 μL of pure K3EDTA (4.4 mmol/L). Changes in parameter values, depicted in percent deviation. Dotted lines indicate an analytical variability ( ± 10%), error bars refer to the standard errors. AST/GOT, aspartate transaminase; ALT/GPT, alanine transaminase.

into the heparinized sample. Because of metal ions, such as calcium or iron, being complexed by EDTA, their measured value would decrease and potassium levels would increase because EDTA in evacuated blood collection tubes is usually deployed as its salt, in particular K3EDTA. These theories arose from the work of Calam and Cooper [3], who observed unexplained elevated potassium and suppressed calcium levels. Furthermore, they concluded that the wrong order of draw with subsequent EDTA contamination caused this finding. Sharratt et  al. [7] found traces of EDTA in routine blood samples with subsequent hyperkalemia, hypocalcemia, hypomagnesemia, and hypozincemia [7]. Also, Davidson [8] emphasized that the interference of EDTA in blood samples affected measured potassium, calcium, magnesium, unsaturated ironbinding capacity, bicarbonate, aspartate transaminase, alanine transaminase, lactate dehydrogenase, creatine kinase, alkaline phosphatase, and α-amylase levels. LimaOliveira et al. [17] report a case in which EDTA potentially altered results for potassium and calcium. Other authors, however, could not find any alterations in laboratory parameters by different orders of draw. Majid et  al. [18]

tested the order of draw theory in 35 patients and 12 controls without finding a significant difference in potassium or calcium levels. Sulaiman et al. [19] investigated a potential EDTA carryover using the Sarstedt Safety Monovette aspiration system. EDTA could not be detected in any of the 10 participants, nor were analytes such as potassium, calcium, magnesium, alkaline phophatase, zinc, or iron significantly altered. These findings were supported by the work of Cornes et al. who evaluated the potential interference of EDTA on potassium, calcium, magnesium, alkaline phophatase, zinc, and creatinine values in 11 subjects [14, 20]. The most recent work in this field is by Salvagno et  al. [21], who investigated EDTA-induced alterations of sodium, potassium, calcium, magnesium, and phosphorus levels and came to the same results as Majid et  al., Sulaiman et al., and Cornes et al. Based on the work of Calam and Cooper [3], a certain order of draw was recommended to prevent an additive carryover: first draw, citrate-containing tube; second draw, heparin-containing tube; third draw, K3EDTA-containing tube; last draw, oxalate-fluonide-containing tube. These recommendations were adopted into the WHO and

Brought to you by | University of Michigan Authenticated Download Date | 5/20/15 1:36 AM

6      Cadamuro et al.: EDTA carryover during blood collection

CLSI guidelines [11, 12], which differ slightly regarding the precise order of draw, yet both guidelines emphasize the collection of heparinized or serum tubes prior to EDTA tubes to prevent EDTA carryover. Because these recommendations are rarely adhered to in clinical practice, preanalytical variabilities in clinical chemistry parameters should be observed far more often. This circumstance and the conflicting evidence in the current literature put the principle of the order of draw into question. Limitations in the previous studies were missing exclusions of other causes leading to hyperkalemia and hypocalcemia. Majid et al. [18] discussed the difficult venipuncture as possible bias, leading to local tissue damage and subsequently inducing release of potassium from damaged cells. This high extracellular potassium leads to a membrane depolarization of local cells, causing calcium influx and resulting in high potassium and low calcium measurements. Another limitation was the small sample sizes of the studies. To our knowledge, no one has yet tried to simulate an EDTA carryover in vitro or investigate the required concentrations of EDTA to significantly alter laboratory parameters. Our work therefore focuses on three main issues: the carryover of EDTA into the subsequent blood collection tube including the amount of transferred EDTA, the EDTA concentrations needed to alter laboratory parameters significantly, and the possibility of EDTA being carried over in “real-life” clinical settings when deviating from current guidelines by collecting blood into a syringe with subsequent tube filling. To test for a carryover of EDTA, we used tandem mass spectrometry analyses as the most sensitive detection method currently available. We simulated blood collection by drawing distilled water into an evacuated EDTA tube and a non-additive tube thereafter. Material from the second tube was analyzed for EDTA contamination. Because the deposition of EDTA differs between the manufacturers of commonly used evacuated blood collection tubes – Becton Dickinson uses liquid K3EDTA, whereas Greiner BioOne applies K3EDTA by vapor deposition – we investigated tubes from both vendors. To exclude random errors, we repeated the experiment 10 times. In all 20 measurements, no EDTA was detectable to a LOD of 0.5 μmol/L. This confirms previous studies, which also failed to detect traces of EDTA in the respective specimens. These publications, however, mainly used a colorimetric assay for EDTA quantification with a higher limit of quantification (LOQ) of 200 μmol/L [14, 19, 22]. At this concentration, some clinical chemical analytes could already be altered, as shown in the second part of our study.

To answer the second question – how much EDTA is needed to significantly alter laboratory parameters – we performed a mixing experiment, adding increasing amounts of EDTA to heparinized blood and subsequently measured sodium, potassium, magnesium, calcium, iron, α-amylase, creatinine kinase, aspartate transaminase, and alanine transaminase levels. EDTA as a hexadentate ligand and chelating agent binds to heavy metal ions such as Fe3+, Hg2+, Cu2+, Pb2+, Ni2+, Zn2+, Cd2+, Co2+, A12+, Fe2+, Mn2+, and alkaline earth metals such as Ca2+ or Mg2+ in descending order of their affinity to EDTA. Alkaline metals, such as sodium or potassium, have the lowest affinity to EDTA and therefore the smallest stability constant. Keeping this in mind, the parameters iron, calcium, and magnesium should decrease when adding EDTA in our experiment. The effect of EDTA on other ions was not investigated in this work because we focused on high-frequency parameters. Depending on the EDTA salt used, the parameters potassium or sodium should increase upon addition of K3EDTA or Na2EDTA, respectively. We divided this experiment into three parts. In part 1 of our mixing experiment, where we determined the amount of EDTA needed to alter test results, calcium, magnesium, and iron values decreased with increasing simulated EDTA whole blood carryover of 10 μL, 100 μL, and 1 mL, respectively. Potassium levels increased as expected; however, the volume of transferred blood needed to significantly increase this parameter was relatively high with 100 μL. All other measured analytes were not altered up to a simulated carryover of 1 mL of EDTA whole blood. To exclude substances other than EDTA that were responsible for parameter alterations, Na2EDTA was used in a second set of experiments. Because all measured alterations in this part of the experiment were nearly identical to those of part 1, we could confirm that EDTA was responsible for the parameter alterations observed. The only differences in measured parameter values between parts 1 and 2 were that we now could observe the effect of EDTA on potassium, which was unaffected up to a simulated carryover of 1 mL EDTA whole blood and that sodium levels increased depending on the amount of added Na2EDTA. Because the amount of EDTA needed to alter clinical chemical test results is way above the LOD of our mass spectrometry assay of 0.5 μmol/L, it seems unlikely that the parameters examined in our work are altered due to an additive carryover. Therefore, our work is in contrast to studies by Calam and Cooper [3], Sharratt et al. [7], or Davidson [8]. However, we can confirm more recent work in the field, such as publications from Cornes et al. [14] or Salvagno et al. [21].

Brought to you by | University of Michigan Authenticated Download Date | 5/20/15 1:36 AM

Cadamuro et al.: EDTA carryover during blood collection      7

Adcock [4] hypothesize that it would be possible to carryover pure EDTA, which adheres to the plunger of the collection tube. To address this aspect, in part 3 of our experiment, we simulated a carryover of pure EDTA by adding 1 μL (final concentration, 183 μmol/L), 5 μL (final concentration, 916 μmol/L), and 10 μL (final concentration, 1833 μmol/L) of K3EDTA into 1  mL of heparinized blood and subsequently measured the same biomarkers as in parts 1 and 2. Calcium and magnesium levels were decreased and potassium levels were already increased, however, at a very low volume of 1-μL carryover of pure K3EDTA. Iron, as well as all the other analytes measured, was not altered up to the addition of 10 μL of EDTA. We could demonstrate that preanalytical alterations of potassium, calcium, and magnesium levels are possible by a potential K3EDTA carryover; however, circumstances contributing to such a preanalytical setting must be given. This in vitro EDTA contamination could potentially occur with open blood collection systems by syringe needle or syringe tip contamination with K3EDTA when delivering collected blood into EDTA sample tubes before other tubes [23]. In a study performed in Sandwell and West Birmingham Hospitals NHS Trust, Berg et al. [13] showed that a significant amount of blood sampling in clinical practice is carried out contrary to current recommendations. Among other techniques, blood was collected into a syringe with subsequent transfer into evacuated sampling tubes either by cap piercing or by removing the cap. These circumstances could explain why EDTA contamination of routine blood samples still can be found [7, 17, 24] even though a carryover of 1 μL of liquid additive from one tube to another is unlikely when using a closed vacuum collection system according to current guidelines. We therefore tried to reproduce such a contamination by comparing vacuum blood collection vs. syringe blood collection with subsequent transfer into EDTA and heparinized tubes in healthy volunteers. Neither in heparin tubes from Greiner BioOne nor from corresponding BD tubes, collected with the vacuum system, EDTA could be detected (LOD = 0.2 μmol/L). However, in heparin samples in which we intentionally dipped an EDTA whole-blood-contaminated needle, while transferring the blood from a syringe to the respective heparin tube, traces of EDTA could be detected. We found a slight difference in the transferred amount of EDTA between the two tube vendors, being slightly higher in those of BD. This circumstance can be explained by the difference in EDTA deposition. Because BD uses liquid EDTA, a carryover into a subsequent tube by needle tip contamination is much more likely in these tubes than in tubes from Greiner BioOne, who spray dry EDTA onto the tube wall. Additionally, we measured all analytes

mentioned above in heparin samples both before and after EDTA tube filling. As expected, we could not find any significant difference when comparing analyte values of the corresponding samples collected by the closed vacuum systems. Also, sample comparison collected by syringe showed no significant difference before and after filling of the EDTA tube in any of the analytes. This finding is in concordance to hypothetical calculations and the results of our mixing experiment, in which much higher EDTA concentrations would be needed to alter values of metal ions. According to the results of this experiment, syringe needle contamination with EDTA whole blood can be ruled out as cause of clinically relevant EDTA contamination, as found by studies surveying routine blood samples from the clinics [7, 17, 24]. However, contamination due to an accidental transfer of pure EDTA during syringe blood collection is still a possible explanation for these carryovers. Another cause could be the EDTA whole blood poured into a heparin tube. Sodium, iron, α-amylase, creatinine kinase, aspartate transaminase, and alanine transaminase levels, however, would not be altered under any EDTA-contaminating circumstances. A limitation of this study is that contamination with additives other than EDTA, such as citrate, heparin, fluoride, or oxalate, were not part of this work. Additionally, we focused on laboratory parameters frequently used in a clinical setting that were described as being susceptible to EDTA contamination in previous studies. Therefore, we cannot exclude an EDTA-associated reduction of metal ions or alkaline metal ions with much lower plasma concentrations than calcium or magnesium. Our observations, together with results from von Salvagno et al., Cornes et al., Majid et al., and Sulaiman et al., suggest that an in vitro interference of the parameters measured in our experiments due to an additive carryover is highly unlikely, even if analyte changes are possible in principle, but only if preanalytical recommendations are severely disregarded. A re-evaluation of current preanalytical recommendations on this topic seems worth considering. Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission. Financial support: None declared. Employment or leadership: None declared. Honorarium: None declared. Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.

Brought to you by | University of Michigan Authenticated Download Date | 5/20/15 1:36 AM

8      Cadamuro et al.: EDTA carryover during blood collection

References 1. Kruse-Jarres JD. Labormedizin 2000 – versuch einer prognose. J Lab Med 1994;18:213–9. 2. Plebani M, Carraro P. Mistakes in a stat laboratory: types and frequency. Clin Chem 1997;43:1348–51. 3. Calam RR, Cooper MH. Recommended “order of draw” for collecting blood specimens into additive-containing tubes. Clin Chem 1982;28:1399. 4. Adcock D. Sample integrity an preanalytical variables. In: Kitchen S, Olson J, Preston F, editors. Quality in laboratory hemostasis and thrombosis. Chichester, West Sussex: Blackwell Publishing, 2009:31–42. 5. Jones JD. Factors that affect clinical laboratory values. J Occup Med 1980;22:316–20. 6. Jacobsen R. Addressing “order of draw” in comparative calcium levels. MLO Med Lab Obs 2001;33:6. 7. Sharratt CL, Gilbert CJ, Cornes MC, Ford C, Gama R. EDTA sample contamination is common and often undetected, putting patients at unnecessary risk of harm. Int J Clin Pract 2009;63:1259–62. 8. Davidson DF. Effects of contamination of blood specimens with liquid potassium-EDTA anticoagulant. Ann Clin Biochem 2002;39:273–80. 9. Schaeffer J, Triplett DA. Case history D: increased APTT due to the presence of heparin. Lab World 1981;32:39–40. 10. Ernst DJ, Calam R. NCCLS simplifies the order of draw: a brief history. MLO Med Lab Obs 2004;36:26–7. 11. World Health Organization. WHO guidelines on drawing blood: best practices in phlebotomy. Geneva, Switzerland: WHO Press, 2010. 12. Institute CaLS. Procedures for the collection of diagnostic blood specimens by venipuncture. Approved standard – 6th ed., vol. H3-A6. Clinical and Laboratory Standards Institute, 2007. 13. Berg JE, Ahee P, Berg JD. Variation in phlebotomy techniques in emergency medicine and the incidence of haemolysed samples. Ann Clin Biochem 2011;48:562–5.

14. Cornes MR, Sulaiman RA, Whitehead SJ, Othonos N, Ford C, Gama R. Incorrect order of draw of blood samples does not cause potassium edta sample contamination. Br J Biomed Sci 2012;69:136–8. 15. Lowe R, Go EP, Tong GC, Voelcker NH, Siuzdak G. Monitoring EDTA and endogenous metabolite biomarkers from serum with mass spectrometry. Spectrosc Int J 2005;19:137–46. 16. Miller ML, McCord BR, Martz R, Budowle B. The analysis of EDTA in dried bloodstains by electrospray LC-MS-MS and ion chromatography. J Anal Toxicol 1997;21:521–8. 17. Lima-Oliveira G, Lippi G, Salvagno GL, Montagnana M, Picheth G, Guidi GC. Incorrect order of draw could be mitigate the patient safety: a phlebotomy management case report. Biochem Med (Zagreb) 2013;23:218–23. 18. Majid A, Heaney DC, Padmanabhan N, Spooner R. The order of draw of blood specimens into additive containing tubes not affect potassium and calcium measurements. J Clin Pathol 1996;49:1019–20. 19. Sulaiman RA, Cornes MP, Whitehead SJ, Othonos N, Ford C, Gama R. Effect of order of draw of blood samples during phlebotomy on routine biochemistry results. J Clin Pathol 2011;64:1019–20. 20. Cornes MP, Ford C, Gama R. The order of draw, myth or science. Clin Chem Lab Med 2013;51:e285. 21. Salvagno G, Lima-Oliveira G, Brocco G, Danese E, Guidi GC, Lippi G. The order of draw: myth or science? Clin Chem Lab Med 2013;51:2281–5. 22. Davidson DF. EDTA analysis on the Roche modular analyser. Ann Clin Biochem 2007;44:294–6. 23. Fitzpatrick MF, Newell J, Grimes H, Egan EL. Spurious increase in plasma potassium concentration and reduction in plasma calcium due to in vitro contamination with liquid potassium edetic acid at phlebotomy. J Clin Pathol 1987;40:588. 24. Davidson DF. A survey of some pre-analytical errors identified from the biochemistry department of a scottish hospital. Scott Med J 2014;59:91–4.

Brought to you by | University of Michigan Authenticated Download Date | 5/20/15 1:36 AM

Relevance of EDTA carryover during blood collection.

The order of draw is regarded as a preanalytical issue to prevent carryover of additives during blood collection. Our objective was to prove the theor...
488KB Sizes 0 Downloads 10 Views