Printer Friendly

The effects of over-anticoagulated blood on hematocrit values by the microcentrifuge method.

ABBREVIATIONS: Aut = automated; CBC = complete blood count; CT = capillary tube; EDTA = ethylenediamine-tetraacetic acid; Hct = hematocrit; Hep = heparinized capillary tube; NonHep = non-heparinized (i.e., no anticoagulant) capillary tube; RBC = red blood cell; RPM = revolutions per minute; WBC = white blood cell.

INDEX TERMS: capillary tube; hematocrit; microhematocrit; microhematocrit centrifuge.

The spun microhematocrit is a method which is often used in screening patients for anemia in physician office laboratories, clinics, and blood donation centers, and as a back-up method in some hospital clinical laboratories. It is simple to perform, inexpensive, and can provide results in about five minutes. In 1993, the FDA added the spun microhematocrit to the list of CLIA-waived tests, thus allowing non-laboratory personnel to perform this procedure. (1) Personnel such as nurses, patient care technicians, medical assistants, and various other non-laboratory personnel are often the ones performing waived tests in point-of-care situations.

There are heparinized capillary tubes available for fingerstick blood samples and non-heparinized tubes for anticoagulated blood samples. Non-laboratory personnel may not always appreciate the different types of capillary tubes and their associated use. It is possible that personnel performing a spun microhematocrit could mistakenly use a heparinized tube with an anticoagulated blood sample. The question then is whether the heparin lining of the capillary tube produces a significant lowering of the hematocrit value due to the additional dilution from an already diluted EDTA-anti-coagulated blood sample. A study to answer this question, which used 30 healthy adults, compared hematocrit values by the spun microhematocrit method using heparinized versus non-heparinized capillary tubes, both filled from an EDTA-anticoagulated Vacutainer[R] tube. Hematocrit values ranged from 35% to 49%. Based on a correlation coefficient of r = .99, the researchers concluded that results from the two types of capillary tubes showed no significant difference in hematocrit result regardless of which type of capillary tube was used with an EDTA specimen. (2) The purpose of the study described herein was to determine if these results could be replicated with a larger sample size which included both well and ill adults, and to expand the methods compared to include an automated hematocrit.

Some factors can affect spun microhematocrit readings, including improper sealing of the capillary tube resulting in blood loss from the tube, not mixing the specimen, the time and speed of centrifugation outside the standard five minutes at 10,000 to 15,000 RPM, incorrect reading with the hematocrit reader, excessive trapped plasma in the packed red cells such as from neonate hyperviscosity syndrome, dehydration of the patient, capillary tube size outside the usual 75mm with 1.1 mm diameter, oxygenation of the blood sample, and whether the patient is sedentary or active before being drawn. (3-9) Both increased anticoagulant and prolonged sitting of red blood cells in an EDTA tube can falsely lower a microhematocrit result due to shrinkage of the RBCs. Poikilocytes such as sickle cells and disorders which affect MCV are also cited as a cause of error. (3) These factors were incorporated into specimen eligibility criteria for the study as described below.


Before beginning this study, the protocol and design were approved by East Carolina University's institutional review board for human subject research. The hematocrit values on each sample were recorded by a unique identification number with no personal or identifying information recorded for the patients. All data was kept in a locked file cabinet except when in use by one of the researchers. All results are reported only in an anonymous, aggregate format.


Patients of Nash General Hospital who had venous blood drawn for a CBC during a three-week period in November 2006 provided the potential blood specimens for the study. Only specimens from patients 19 years of age or older were accepted for the study. Each specimen was drawn in a three ml Vacutainer [R] tube with dry K2 EDTA anticoagulant by laboratory phlebotomy staff as part of the early morning test orders. The rejection criteria for the study specimens are displayed in Table 1. A total of 101 specimens were accepted for inclusion in the study.


Prior to beginning the specimen analyses, proper quality control for the microcentrifuge method was performed including timer and RPM verifications. Appropriate controls were run on the Sysmex XE-2100 each day that specimens were collected. The capillary tubes used were all made by Drummond, 75 mm height and 1.1 mm diameter, and plastic and mylar wrapped for safety. It should be noted that the Sysmex XE-2100 performs a measured hematocrit based on cumulative RBC pulse height, which is different than some other instruments that calculate the hematocrit based on the MCV and RBC count.

The hematocrit from the Sysmex XE-2100 was recorded as the automated method result for each accepted specimen. These specimens were next prepared for manual analysis by microcentrifuge. Each EDTA tube was inverted five times for adequate mixing. Two heparinized capillary tubes and two non-heparinized capillary tubes were filled until two-thirds full with blood sample, and one end was sealed with Crito-seal. Tubes were placed in a Damon/IEC Division microhematocrit centrifuge in opposite slots with the sealed ends toward the outside and centrifuged for five minutes at 13,000 rpms. Once the microhematocrit centrifuge had completely stopped, without use of the brake, the hematocrit was read within 10 minutes as the percent of whole blood occupied by packed RBCs using a Damon/IEC Division Microcapillary Reader. In reading the hematocrit value, care was placed on excluding the buffy coat (i.e., layer of platelets and leukocytes) and reading the nonheparinized tubes and corresponding heparinized tubes in pairs to avoid sample confusion. The readings for each type of capillary tube were performed in duplicate, read once by two researchers, and averages were taken. The duplicate readings for each type of capillary tube had to agree within one percent to be acceptable.

Data analysis

All data entry and statistical analyses were performed in SPSS-PC+ version 15.0. The data include hematocrit values for each of the three methods as well as calculated variables representing the difference between each method with each of the other two methods. Descriptive statistics were calculated for all three methods (automated, non-heparinized capillary tubes, heparinized capillary tubes) as well as for method differences. Pearson (pairwise) correlation coefficients were calculated by bivariate analyses and scatterplots produced comparing each method with each of the other two methods, along with a repeated one-way analysis of variance (ANOVA) for all three methods. All analyses were evaluated for significance at an alpha of .01. For the two types of capillary tubes, analyses were performed only on the hematocrit averages for the duplicate readings.


The hematocrit values for all three methods ranged from 20.0% to 53.8%, with a mean of 35.7% for all three methods combined. Descriptive statistics for each method are shown in Table 2. Comparisons of methods by scatterplots are in Figures 1 through 3. All three scatterplots show only minimal differing of results between methods with the hematocrits of the two types of capillary tubes agreeing the most closely.

Review of the descriptive statistics for each method as well as for differences in means between methods indicates that the three methods produced similar results with differences that would not be clinically significant for patient care. Over half of the samples produced hematocrit values with zero difference when comparing results of the two types of capillary tubes, though only two samples had zero differences when the automated results were compared to each type of capillary tube. For 68% of the samples, the automated hematocrit was higher than the microhematocrit centrifugation values from either of the two types of capillary tubes. The two capillary tubes had a smaller difference in means than did either capillary tube compared to the automated method (Table 2). All Pearson (pairwise) correlation coefficients comparing two methods were statistically significant at <.01 as follows: 1. for automated vs. heparinized capillary tubes, r(99) = .980, p < .005; 2. for automated vs. non-heparinized capillary tubes, r(99) = .981, p < .005; and 3. for heparinized vs. non-heparinized capillary tubes, r(99) = .996, p < .005.

When the repeated measures one-way ANOVA was performed, the variability about the means was found to be significantly different between the methods with an F(2,100)=27.8,p<.005. Though the differences in hematocrit between methods appear not to be clinically significant, there was probably enough variability about the means to be statistically significant, particularly with a large sample where small differences can cause statistical significance. So, though the results vary together in a positive relationship, there may be relatively large differences between methods for some specimens in this study.


As in previous research studies, (2) a statistically significant correlation (p < .005) between the hematocrit results by the two types of capillary tubes was found when filled with whole blood from an EDTA-anticoagulated Vacutainer[R] tube. Using the Sysmex XE-2100 hematology analyzer, statistically significant correlations (both p < .005) were also found with automated hematocrit results when compared to each of the two types of capillary tubes. Based on repeated measures one-way ANOVA, a significant difference between the three methods based on variability about the means (p < .005) was also found. Upon review of the differences in hematocrit values between the two types of capillary tubes and between the automated method versus each type of capillary tube, the differences were statistically significant based on the large sample size as used in this study but were still minimal enough not to alter the direction of patient evaluation relative to anemia screening by hematocrit. The differences between the methods appear clinically insignificant with the closest agreement being between the hematocrit values produced by the two types of capillary tubes when filled from an EDTA tube.


Interestingly, it has long been held that automated hematocrit results will be one percent to three percent less than the microhematocrit results when abnormal cells such as sickle cells, macrocytes, hypochromic cells, and spherocytes are present because of the effect of trapped plasma in the micro-centrifuge method. (3) The results from this study, though, found for 68% of the specimens, the automated hematocrits were higher than the microhematocrits with either type of capillary tube. This may be due to the fact that we did not accept specimens for this study that had flags, as well as some other extreme results on the automated report which may have eliminated blood samples with abnormal cells. Also, the difference between automated and microcentrifuge hematocrit results stated above are more applicable to analyzers that calculate the hematocrit, (3) whereas the analyzer used in this study measures the hematocrit directly.



The results of this study reinforce previous research (2) and support the belief that it does not make a clinical difference in the hematocrit result if non-laboratory personnel in point-of-care testing situations were to use a heparinized, as opposed to the recommended non-heparinized, capillary tube to fill it from an EDTA-anticoagulated tube and then perform a microhematocrit. In essence, use of the non-recommended capillary tube would not produce a clinically significant difference in results such that patient evaluation for anemia by hematocrit would not be affected by the procedural error.

The study conclusions are limited by the study design for specimen acceptance which created an ideal situation not reflective of busy clinical laboratories. Non-laboratory personnel performing microhematocrits will not be able to screen samples as was done in this study. These conclusions are further limited to use of the same materials and methods employed for this study. Additional research should investigate this topic but in an actual clinic or physician office laboratory staffed by non-laboratory personnel to see if similar results are obtained under true working conditions.


Clin Lab Sci encourages readers to respond with thoughts, questions, or comments regarding this article. Email responses to In the subject line, please type "CLINLAB SCI 21(3) RR BAMBERG" Selected responses will appear in the Dialogue and Discussion section in a future issue. Responses may be edited for length and clarity. We look forward to hearing from you.

Acknowledgements: Thanks are expressed to Charlotte Mizelle and Randy Duke in the Clinical Laboratory at Nash General Hospital, Nash Health Care System, Inc., Rocky Mount NC for providing blood samples for this study and allowing the researchers to perform the hematocrit testing in the Clinical Laboratory on their Damon/IEC Division microhematocrit centrifuge and microcapillary reader.

This study was presented as a platform presentation at the 2007 Carolinas Clinical Connection conference, though no written proceedings of abstracts or papers were produced.


(1.) US Food and Drug Administration. CLIA--Clinical Laboratory Improvement Amendments, Waived Tests. Available from http:// Accessed 2007 Oct 22.

(2.) Hughes VC, Cox R, Ernest C, and others. Choosing heparinized over non-heparinized capillary tubes in the centrifugal microhematocrit method. Lab Med 2005;36(9):556-7.

(3.) Hippel TG. Routine testing inhematology. In Rodak BF, Fritsma GA, Doig K, editors. Hematology: clinical principles and applications, 3'.d edition. St. Louis, MO: Saunders/Elsevier; 2007:160-74.

(4.) Bull BS, Caucho VPR, Hay KL. Control of analyzer slope and intercept in the measurement of packed red cell volume (PCV), part 1. Blood Cells, Molecules, and Diseases 2002;28(2):108-15.

(5.) Dosik H, Prasad B. Coulter S hematocrit and microhematocrit in polycythemic patients. Am J Hematol 1978;5(1):51-4.

(6.) Penn D, Williams PR, Dutcher TF, Adair RM. Comparison of hematocrit determinations by microhematocrit and electronic particle counter. Am J Clin Path 1979;72(1):71-4.

(7.) Mayer GA Diurnal, postural and postprandial variations of hematocrit. Canadian Med Assoc J 1965;93(19):1006-8.

(8.) Bryner MA, Houwen B, Westengard J, Mein O. The spun microhaematocrit and mean red cell volume are affected by changes in the oxygenation state of red blood cells. Clin and Lab Haematol 1997;19(2):99-103.

(9.) Solomon HM, Grindon AJ. The effect of capillary tube diameter on microhematocrit value. Transfusion 1986;26(2):199-202.

The peer-reviewed Research and Reports Section seeks to publish reports of original research related to the clinical laboratory or one or more subspecialties, as well as information on important clinical laboratory-related topics such as technological, clinical, and experimental advances and innovations. Literature reviews are also included. Direct all inquiries to David L McGlasson MS CLS(NCA), 59th Clinical Research Division/SGRL, 2200 Berquist Dr., Bldg. 4430, Lackland AFB TX 78236-9908,

Richard Bamberg PhD MT(ASCP)SH CLDir(NCA) CHES is professor and chair, Department of Clinical Laboratory Science, College of Allied Health Sciences, East Carolina University Greenville NC.

Thomas Gwyn MT(ASCP) is medical technologist, Microbiology Section, Department of Pathology/Laboratory, Pitt County Memorial Hospital Greenville NC.

Jason Miller MT(ASCP), Maurice Thompson MT(ASCP), and Phyllis Transou MT(ASCP) are medical technologists, Clinical Laboratory, Nash Health Care System, Inc., Rocky Mount NC.

Address for correspondence: Richard Bamberg PhD CLDir(NCA), professor and chair, Department of Clinical Laboratory Science, College of Allied Health Sciences, Health Science Building 3410, East Carolina University, Greenville, NC 27858-4353. (252) 744-6060, (252)744-6068 (fax).
Table 1. Rejection criteria for blood specimens

Based on visual inspection of the specimen and
test(s) order collection time:

< 1.0 ml blood in the EDTA tube

> 1.0 hours since time of collection

Presence of hemolysis, lipemia, icterus, or clots

Based on CBC results from the Sysmex-2100:
Hematocrit > 55.0%

Hemoglobin > 18.0 or < 7.0 g/dl

WBC count > 30,000 or < 2,000/ul

MCHC > 36%

Instrument report had any flag(s) noted

A manual differential and morphology was performed
for any other reason

Table 2. Descriptive statistics by methods (N = 101)

Hematocrit Capillary tube

(in %) Automated Heparinized Non-heparinized

Mean 36.16 35.61 35.40
Median 36.00 35.00 35.00
S.D. 6.02 5.44 5.47
COPYRIGHT 2008 American Society for Clinical Laboratory Science
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2008 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Bamberg, Richard; Gwyn, Thomas; Miller, Jason; Thompson, Maurice; Transou, Phyllis
Publication:Clinical Laboratory Science
Geographic Code:1USA
Date:Jun 22, 2008
Previous Article:The doctorate in clinical laboratory science: an executive summary.
Next Article:Growth inhibition of human colon cancer cells by plant compounds.

Related Articles
Effects of Different Hematocrit Levels on Glucose Measurements With Handheld Meters for Point-of-Care Testing.
High blood viscosity may be new risk factor for type 2 diabetes. (43% Higher Relative Risk).
Minimum blood volume for CBC tubes.
Fibrinogen levels in Polycythemia Vera patients and hematocrit target for performing a therapeutic phlebotomy in southwest Alabama.
Hb and HCT do not match in oncology patients.
Hematocrit pre-analytical errors.
Development of a Micro-ESR system with potential for in-home use.
Assessment of a portable clinical blood analyzer during space flight.
Analyzer for use in blood centers.
Manual placenta removal may not up blood loss.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters