The present study was an observational and comparative study conducted at the department of pathology, Chettinad Hospital and Research Institute, Tamil Nadu for a duration of 6 months from January 2019 to June 2019. The study was carried out after obtaining approval from the institutional human ethics committee (278/IHEC/1-19). A total of 100 paired samples including both healthy individuals and patients willing to participate in the study between the age groups of 18-65 years were included. The recommendation of Bland Altman [18] was as followed to obtain sample size of 100 for looking the level of agreement between two clinical methods. All patients except pediatric and geriatric cases for whom CBC was requested were included in the study. The volunteers were explained the purpose of this research and an informed consent was obtained. Paediatric, geriatric and pregnant patients were excluded from the study.

From each volunteer two venous blood samples were taken from a single cubital region by standard venipuncture protocol and were collected into two spray dried K2EDTA vacutainers (BD vacutainer). A total of 4 mL blood was drawn from each individual. In one vacutainer 3 mL of blood was drawn and in another vacutainer 1 mL of blood was drawn. Level of blood filling was marked externally on the vacutainers. The one marked at 1 mL was labelled as A (underfilled EDTA), other one marked at 3 mL waslabelled as B (standard). Each vacutainer was then inverted to 180° for 5-6 times to ensure proper mixing of anticoagulant and blood.

Samples were analysed using BECKMAN COULTER LH 780 automated analyser (Beckman coulter, India). Blood samples were analysed on the same day. All the samples were analysed only when the quality controls were within acceptable limits. Haemolysed, grossly contaminated and clotted samples were rejected. The values obtained were entered in the spreadsheet. The coulter method analyses was based on electrical impedance and Volume Conductance Scatter (VCS) technology. RBC count, WBC total count and platelet count were estimated by electrical impedance. Other RBC parameters such as MCV, MCH, MCHC, RDW were calculated. WBC differential analysis was based on VCS technology. Haemoglobin was estimated by cyanmethaemoglobin method; following WBC count the lysed WBC dilution drains into haemoglobin cuvette. A beam of white light from an incandescent lamp goes through the cuvette and through an optical filter that had a transmission wavelength of 525 nm.

Collected data was statistically analysed using SPSS software v.18. Standard deviation, standard error, mean difference and 95% confidence interval were calculated. Paired student’s t-test was used to analyse the results. The data from under-filled EDTA vacutainers were compared with the standard collection. The p-value <0.05 was considered statistically significant.

The comparison of the results between under-filled 1 mL K2EDTA vacutainers and standard 3 mL K2EDTA vacutainer are given in [Table/Fig-1].

Hundred samples of under-filled vacutainers were compared with the standard volume. RBC parameters such as RBC count, HBG, MCV, MCH, MCHC, RDW, HCT; WBC parameters such as total count and differential count; platelet count and MPV were comparable by statistical analysis between both under-filled vacutainers and standard volume [Table/Fig-1].

The result of this study shows that all the parameters in the CBC showed comparable results by statistical analysis between under-filled K2EDTA vacutainers (1 mL) and standard 3 mL vacutainers. There was no significant variation in statistics even when the blood volume was as low as 1 mL which is 67% less than that of the recommended volume.

Blood samples collected in both K2- and K3EDTA can be stored at 4°C without significant alteration of blood parameters for 24 hours [11]. K2EDTA is preferred over K3EDTA because it is more hypertonic and causes erythrocyte dehydration and shrinkage [19]. Another drawback of K3 EDTA is that as it is a liquid anticoagulant it results in the dilution of the blood specimen. It lowers most of the measurements including RBC, WBC, platelet, MCV and Haemoglobin [19]. Underfilling of vacutainers leads to an excess of EDTA, which can cause platelet disintegration or abnormally enlarged platelets resulting in spurious thrombocytopenia or thrombocytosis. It can also alter RBC and WBC morphology. Previous studies done on the use of underfilled K2EDTA vacutainers for CBC analysis showed contradicting results. ForLH 700 series, the instrument aspirates only 0.3 mL of sample in automatic aspiration mode and only 0.2 mL by manual aspiration mode. However, at least 1 mL of sample with proper proportion of anticoagulant is required for automatic haematology analysers as per guidelines [20].

Only few studies have been done on underfilled K2EDTA vacutainers and the results of our study are in line with some of the earlier studies [9,16]. In the present study, the results obtained shows that underfilling upto 1 mL does not alter the haematological parameters and showed comparable results with recommended volume.

A study by Gupta V et al., showed that underfilling of K2 EDTA vacutainers upto 1 mL did not affect the haematological parameters in healthy individuals [9]. These results are in concordance with the present study. In addition, our study population also included blood samples obtained from patients along with healthy individuals. Hence this study also proves that the CBC results obtained from underfilled K2EDTA vacutainers are comparable between healthy and diseased individuals.

Xu M et al., compared standard 4 mL to underfilled 2 mL, 1 mL and 0.5 mL K2EDTA vacutainer on healthy blood donors using Sysmex XE-2100 [16]. This study showed that upto 1 ml, all the parameters were comparable to the standard 4 mL excluding platelet count, lymphocyte percentage and automated reticulocyte count. However, when underfilling is done upto 0.5 mL, only the RDW, MCV and neutrophil percentage were comparable, all the other parameters showed significant variation. The present study showed comparable results in all parameters upto 1 mL. However, reticulocyte count and underfilling <1 mL was not included in the study.

A study by Pan L et al., compared the standard 3 mL sample to underfilled K2EDTA of 1 mL, 0.6 mL and 0.3 mL in various clinical settings [17]. This study showed that when the blood sample volume was reduced from 1 mL to 0.3 mL, there was a significant increase in the variation of haematological parameters like RBC, Hb, HCT and MCV. In contrast our study showed no significant variation between 3 mL and 1 mL sample.

The limitations in the present study include that only underfilling upto 1 mL was compared and only one type of haematology analyser was used. Also, the present study did not compare the effect of excess anticoagulant on the cell morphology. Blood samples from paediatric patients in whom underfilling of vacutainer is more commonly encountered was not included in this study.

The present study did not show any significant variation in the parameters between the underfilled K2EDTA vacutainers (1 mL) and the standard volume for CBC analysis using automated analyser. Therefore, the CBC results generated from the under-filled vacutainers can be accepted for clinical diagnosis in conditions where repeat phlebotomy is a concern or impossible. This will reduce the delay in reporting the results caused due to rejection of under-filled K2-EDTA vacutainers without affecting the results. However, more studies of this type needs to be replicated in other laboratories with various other automated analysers in healthy as well as various pathological conditions in order to revise the current CLSI guidelines which are mainly based on liquid K3EDTA vacutainers.