Agreement between blood draw techniques for assessing platelet activation by flow cytometry

It is widely believed that assays of platelet activation are susceptible to pre-analytical variables related to blood draw technique. We assessed platelet activation by whole blood flow cytometry and investigated the effects of: 1) drawing blood into vacuum tubes or manually-aspirated syringes, and 2) discarding the first drawn blood sample (discard tube). Platelet P-selectin expression and platelet-monocyte complexes were measured by flow cytometry under both basal conditions and following stimulation with 0.1, 1 or 10µM ADP. Bland-Altman plots demonstrated agreement between results for vacuum tube and syringe-aspirated samples with an a priori -defined clinically relevant agreement limit of 5%. Agreement of results was also observed between discard tube and second draw samples for both vacuum-driven and manually aspirated blood. We conclude that a vacuum tube or a manually-aspirated syringe can be used when assessing platelet activation by flow cytometry and that there is no need for a discard tube. The present study represents the first attempt to apply agreement statistics to pre-analytical determinants of platelet activation assessed by flow cytometry. We have shown that the initial blood drawn, or subsequent samples drawn by butterfly needle into a vacuum tube or manual syringe, did not influence platelet activation assessed by flow cytometry.


Introduction
Unresponsiveness or hyper-reactivity of platelets can lead to disease, and assays of platelet function are commonly used diagnostically and experimentally. Since platelets are prone to become artificially activated ex vivo, rigorous attention is given to pre-analytical variables [1], particularly the technique used for blood collection [2]. Guidelines for the diagnostic evaluation of platelet disorders recommend either vacuum-or syringe-driven aspiration of blood into primary collection tubes [3]. However, many authorities believe that vacuum systems subject blood to higher levels of shear stress, resulting in artificial platelet activation [4]. Furthermore, the Clinical Laboratory Standards Institute (CLSI) recommends discarding the first drawn tube when blood is taken through a butterfly needle [5]. The rationale is that dead space in the tubing leads to underfilling of the primary collection tube leading to an incorrect anticoagulant/blood ratio [6,7].
Flow cytometry allows rapid ex vivo measurement of platelet activation through the assessment of surface receptors such as P-selectin, platelet-monocyte complexes, or fibrinogen binding [8]. Upon platelet stimulation, P-selectin from α-granules is rapidly translocated to the platelet membrane where it can be detected using labelled antibodies.
Surface P-selectin mediates the rolling of platelets on endothelial cells, facilitates interactions with monocytes, and enables the initial platelet-platelet contacts needed for platelet aggregation [9]. P-selectin expression is often used as the 'gold standard' for assessing platelet activation, yet it can be rapidly shed from the platelet membrane so functional assays such as measurement of platelet-monocyte complexes are also performed [10].
The effect of pre-analytical variables on platelet activation assessed by flow cytometry has only been partially evaluated [11]. We, therefore, investigated the effect of blood draw technique on basal and stimulated P-selectin expression and platelet-monocyte complexes measured by whole blood flow cytometry. We applied statistical methods for assessing agreement between different techniques.

Patients and specimens
The study population comprised ten healthy volunteers (four males and six females; mean age 26), recruited from University staff who had not taken platelet-altering medication for a minimum of 14 days. The study was approved by the Ethics Committee of Hull York Medical School. Two consecutive blood samples (Vac-1 and Vac-2) were drawn from the from the median cubital vein in the right arm using a 21g butterfly needle (Becton Dickinson, UK) into 2.7 ml Vacutainers® containing 0.109 mol/l buffered sodium citrate (Becton Dickinson, UK).
Immediately, two consecutive 2.7 ml samples (Syr-1 and Syr-2) were drawn from the opposing arm by manual aspiration into 5.0 ml syringes preloaded with 0.109 mol/l buffered sodium citrate. Other variables including tourniquet tightness and rest time were standardised.
Samples were processed within 20 minutes of venepuncture and flow cytometry performed within 3 hours.

Laboratory tests
40µl of whole blood was incubated at room temperature for 20 minutes in 40µl phosphate- to confirm that the gated population was platelets. Samples were fixed with 400µL of 1% paraformaldehyde before flow cytometric analysis. Platelets were identified by their forward and side scatter properties and CD42b expression. 10,000 platelet events were acquired. We recorded the mean fluorescence intensity (MFI) and the percentage of platelets expressing CD62P with the positive region marker set at the 99 th percentile of the isotype antibody fluorescence.
Measurements are reported as mean and 95% confidence internal (95% CI). Comparisons for Syr-1 vs Syr-2, Vac-1 vs Vac-2, Syr-1 vs Vac-1 and Syr-2 vs Vac-2 were performed using Pearson's correlation alongside Passing and Bablock regression. The Shapiro Wilk test confirmed normality of the differences between each of the pairs. Paired students t-tests assessed the significance of differences between samples and Bonferroni correction was employed for multiple testing. Bland-Altman plots were generated to assess agreement between the pairs for both platelet-monocyte complexes and platelet P-selectin expression [12]. The a priori acceptable agreement between methods was set at 5%. Further explanation of the statistical approach can be found in the supplementary information.

Results
Stimulation with ADP caused concentration-dependent increases in platelet-monocyte complexes and platelet P-selectin expression (Fig. S1). When comparing the different blood  (Table S1). Differences between P-selectin expression (percent positive) and percentage platelet-monocyte complexes between samples were not statistically significant at any ADP concentration (Table I). Furthermore, no statistical significance was seen between P-selectin MFI at any ADP concentration (Table S2). When assessing agreement between methods using Bland-Altman plots, good agreement was demonstrated by the 95% limits of agreement (LOA) and the a priori 5% agreement limit for platelet-monocyte complexes and platelet P-selectin expression for all comparisons (Fig 1).
Higher variability was seen between methods at higher levels of ADP stimulation (proportional bias) but was bidirectional and within acceptable limits (Fig 1).

Discussion
Evidence suggests that the pre-analytical stage is most vulnerable to laboratory error [13].
Although guidelines exist for the standardisation of pre-analytical variables when conducting platelet assays, recommendations are based on light-transmission aggregometry and the platelet function analyser-100 with no data arising from flow cytometric studies [3].
Bland-Altman plots allow systematic differences to be detected between two measurement techniques [12]. Studies investigating the influence of pre-analytical variables on platelet assays suffer from inappropriate statistical analysis [6,14]. Moreover, failure to define a priori clinically meaningful differences hampers interpretation of Bland-Altman plots [15,16]. In the present study, we display Bland-Altman plots displaying a priori clinically meaningful differences and data-defined LOA.
Guidelines advocate that a discard tube of blood is needed when a butterfly device is used [5]. We demonstrate agreement in basal and stimulated platelet P-selectin expression and platelet-monocyte complexes between the first and second collection tubes. We conclude it is unnecessary to discard the first tube of blood drawn.
Agreement was also established between results of blood aspiration via syringe or vacuum tube, which is consistent with guidance that blood may be drawn by either method [3].
The present study represents the first attempt to apply agreement statistics to pre-analytical determinants of platelet activation assessed by flow cytometry. We have shown that the initial blood drawn, or subsequent samples drawn by butterfly needle into a vacuum tube or manual syringe, did not influence platelet activation assessed by flow cytometry. Table Legend   Table I.

Statistical Approach
To assess agreement a stepwise approach was undertaken. Although it is assumed that two techniques for measuring the same output will be closely related, Pearson's correlation coefficient (PCC) was calculated to clarify the presence of a linear relationship between Syr-1-Syr-2, Vac-1-Vac-2, Syr-1-Vac-1, Syr-2-Vac-2. PCC is highly sensitive to the range of values and lacks information about systematic difference therefore PCC does not assess agreement as a high degree of correlation is possible when agreement is poor [1]. Passing and Bablock regression analysis was then undertaken on the above pairs. This is a non-parametric linear regression procedure which is non-sensitive to outliers and fits the parameters of a and b in the linear equation y = a + bx. This reveals constant (regression line intercept a) and proportional (regression line slope b) difference with confidence intervals of 95% (95% Cis).
Therefore, if the 95% Cis for a include zero one can conclude that there is no constant difference between methods. Additionally, if the 95% Cis for b include the value one, then it can be concluded that there is no proportion difference between methods. Overall this allows the for the assumption that x = y and agreement between methods to be presumed. The primary fallacy with the Passing and Bablock regression model is that it derives the agreement of two methods from the data and neglects whether this is within clinically relevant parameters [2]. Therefore, Bland-Altman analysis was undertaken (explained in the main text) to assess whether the agreement between the above pairs was within an a priori 5% agreement limit. With this analysis, a paired students t-test was also computed testing the null hypothesis H0 that the mean of the differences between the results does not differ from 0, against the alternative Ha that it does. Finally, Bonferroni correction was applied to adjust the significant p value for paired students t-tests to account for multiple comparisons. Tables   Table SI. Results of Passing and Bablock regression analysis for Syr-1 vs Syr-2, Vac-1 vs Vac-2, Syr-1 vs