Accuracy and Precision Comparison of the Hemocytometer

and Automated Cell Counting Methods

Laurel R. Stone, Diane R. Gray, Krissy L. Remple, Matthew P. Beaudet
Invitrogen Corporation • 5791 Van Allen Way • Carlsbad, California 92008 • USA

Abstract
The hemocytometer is the most widely used
device for determining cell concentrations,
requiring consistent criteria and tenacity to
obtain measurements correctly
reproducibly. As an alternative, the Countess™
Automated Cell Counter employs digital
Figure 1 – Counting variation Figure 2 – Comparison of inter- and intra- Figure 3 – Automated cell counting viability determination Results
between traditional glass and user error for both manual and across multiple cell types
The experiments shown were used to compare
disposable plastic hemocytometers automated cell counting methods the performance of manual cell counting using a
Viability Comparison hemocytometer to the CountessTM Automated
Hemocytometer Counting Comparison
and Inter- User Comparison Cell Counter.
100
Primary Cells:
1.40E+06 Bead Standard
1.40E+06 Adipocyte •Accuracy of the instruments was determined by

Average Concentration (beads/mL)

1.20E+06 Glass-Hausser
1.20E+06 HASMC performing several counts of the same sample.
imaging and an image analysis algorithm to

Countess Instrument Viability

HPAEC
The CountessTM instrument had a smaller
Concentration (beads/mL)
C-Chip Disposable 80
1.00E+06
identify and enumerate cells in a sample. To Hemocytometer 1.00E+06 Keratinocyte
Adipocyte
standard deviation for both count and viability
Adherent
HASMCCultured Cells:
compare the performance of these two 8.00E+05 8.00E+05
HPAEC
A431 data.
methods, the accuracy, precision, and effective 6.00E+05 6.00E+05
60
Keratinocyte
C2C12
ranges were determined for total cell count and A431
COLO-205 •The CountessTM instrument has an accurate
percent viability. First, using a standardized
4.00E+05 4.00E+05 C2C12
MMM range to 4.0x106 cells/mL and can count to
COLO-205
bead solution, the accuracy and precision of 2.00E+05 2.00E+05
40
MRC-5
MMM 1.0x107 cells/mL, while manual counting
Suspension
MRC-5 Cultured Cells:
glass and disposable hemocytometers were 0.00E+00 methods are limited to approximately 1.0x106
0.00E+00 K562
measured and found to be 30% and 5% Instrument
CountessTM Instrument Glass Hemocytometer Disposable
20
K562
cells/mL.
Hemocytometer
different from the anticipated value, A C-Chip disposable hemocytometer (Manufacturer Instrument Used
•Viability is easily measured by the CountessTM
respectively. Subsequently, disposable Digital Bio Technology) contains two separate, fixed- instrument by using trypan blue dye exclusion.
hemocytometers and the automated counter volume chambers. This is much more consistent than To test the inter-user variation for manual and automated 0
Viability measurements for both manual and
were used to determine total cell numbers and
0 20 40 60 80 100
the glass hemocytometer which relies upon the cell counting methods for multiple users, three separate automated cell counting agree.
percent viability for multiple cell types over a surface tension of the sample to hold the loose cover CountessTM slide chambers were loaded with a standard
Hemocytometer Viability

range of cell concentrations. Using a Student’s
t-test, the results showed the CountessTM
instrument measured cell concentrations and
viability as accurately and precisely as the
disposable hemocytometer. Additionally, the
effective concentration range for the
CountessTM instrument was two times greater
than the hemocytometer, and the viability
range matched the hemocytometer. The
CountessTM Automated Cell Counter overall
produces results much more rapidly without
the problem of operator tedium and fatigue or
compromised accuracy and precision.
bead sample by three different, trained users. Each •Inter-user variation of total cell count was
slip at the correct depth. Three preparations and This experiment was used to judge the accuracy of the CountessTM instrument
chamber was counted by each of the users. The same shown to be less than 1% using the CountessTM
counts were performed for each type using a viability readings. Ten cell lines were counted on the Countess™ Automated Cell
work flow was followed for the hemocytometer counts. instrument and as much as 5.5% for trained
standard latex bead sample (Coulter bead standard, Counter – 4 primary lines (keratinocyte, HPAEC, HASMC, and adipocyte), 5 adherent
The error bars show differences in user count results, users counting manually.
9.57x105±10%) following the manufacturer’s cell lines (C2C12, A431, MRC-5, MMM, and COLO-205), and 1 cultured suspension
suggested protocol. The results show a difference of while the column heights show differences in sample line (K562). For each cell line five samples were prepared using different ratios of Conclusions
30% and 5% from the expected value for the glass preparation for each instrument. As shown in the chart live cells and heat killed cells, to represent theoretical viability points ranging from
and disposable hemocytometers, respectively. Error above, the variation in total bead counts between users is 0% to 100% live cells. One chamber slide was prepared for each sample and
bars show one standard deviation above and below as much as 5.5% when using the hemocytometer, and Due to the critical nature of accurate cell
counted three times each using the CountessTM instrument. Counts of ten squares
the mean concentration for each hemocytometer less than 1% when using the Countess™ Automated Cell counts in biochemical and cellular assays, it is
using disposable hemocytometers were performed for each sample.
type. These data, combined with ease of use Counter. Sample preparation difference was approximately important that the cell counts be easy to
comparisons, resulted in the use of only the 12% for the CountessTM instrument, and approximately perform, be accurate, and give reproducible
disposable hemocytometer in subsequent 15% for each of the hemocytometers. results. We have shown that when compared
hemocytometer comparison experiments. to manual counting using a disposable
hemocytometer, the CountessTM instrument
Figure 4- Accuracy and Precision- Automated cell counting Figure 5-Effective concentration range-Automated cell counting compared to a gives more precise and as accurate results. The
compared to a traditional hemocytometer traditional hemocytometer concentration range of the CountessTM
instrument extends much higher than what a
Counting Comparison- Keratinocytes Viability Comparison- Keratinocytes A serial dilution series of SF-9 cells hemocytometer user could count by eye. This
A B Dilution Series: SF-9 Cells
was made starting with a fresh removes the need for dilutions prior to
Countess
1.20E+06 Countess 99.00
Instrument 4.00E+06 Countess 1 highly concentrated cell sample. counting, thereby improving work flow. Cell
Instrument
1.00E+06 Hemocytometer 97.00 Hemocytometer Countess 2 Three CountessTM slide chambers viability readings were shown to be equivalent
Concentration (cells/mL)

95.00 Measured Concentration (cells/mL) 3.50E+06 Countess 3 were counted three times each for to those determined visually and quantified
8.00E+05 Hemocytometer 1 all of the samples in the dilution using a hemocytometer. Inter-user variation is
% Viability

3.00E+06
6.00E+05
93.00
Hemocytometer 2 series. A manual count of ten also markedly decreased by using the
91.00
2.50E+06
Hemocytometer 3 squares was performed using a CountessTM instrument. These experiments
4.00E+05
89.00 disposable hemocytometer three were performed using over 25 different cell
2.00E+05 87.00
2.00E+06
1.20E+06
times for each sample below 1x106 types, including primary cell lines, with similar
1.00E+06 cells/mL. Due to the density of the results. The Countess™ automated cell counter
1.50E+06
0.00E+00 85.00 8.00E+05 cells, it was difficult to manually streamlines the enumeration of cells to obtain
Instrument Instrument
1.00E+06
6.00E+05
count the samples using the quicker and more reliable results which reduces
The Countess™ Automated Cell Counter hemocytometer for cell
Three separate Countess™ counting chambers were loaded with primary keratinocyte user fatigue and tedium. As is evident in these
4.00E+05

uses the trypan blue dye exclusion method concentrations above 1x106 experiments, automated cell counting improves
2.00E+05
cell samples and counted three times each for a total of nine measurements. These 5.00E+05
to calculate cell concentration and provide 0.00E+00

cells/mL. The CountessTM instrument

results were compared to three counts of ten squares each in a disposable 0.00E+00 2.00E+05 4.00E+05 6.00E+05 8.00E+05
results by providing consistent and accurate
viability information. 0.00E+00
was capable of accurately counting
hemocytometer. Using the standard trypan blue exclusion method and following the 0.00E+00 5.00E+05 1.00E+06 1.50E+06 2.00E+06 2.50E+06 3.00E+06 3.50E+06 4.00E+06 data which is important for reproducibility and
www.invitrogen.com/countess manufacturers protocol, count and viability were determined for both instruments. cells at a much higher concentration statistical confidence.
Theoretical Sample Concentration (cells/mL)
Student’s t-test was used to show that the mean of the two groups are equal for a than was possible using a
95% confidence interval. Error bars show one standard deviation above and below hemocytometer, which reduced the
the calculated mean total cell concentration based on the results obtained from each need for dilution of the sample
instrument. before counting and improved the
workflow.