Hemoglobin Point-of-Care Testing: The HemoCue System 457560

Transcription

Hemoglobin Point-of-Care Testing: The HemoCue System 457560
457560
of Laboratory AutomationSanchis-Gomar et al.
2012
JLAXXX10.1177/2211068212457560Journal
Hemoglobin Point-of-Care Testing: The HemoCue System
Journal of Laboratory Automation
XX(X) 1­–8
© 2012 Society for Laboratory
Automation and Screening
DOI: 10.1177/2211068212457560
http://jala.sagepub.com
Fabian Sanchis-Gomar1, José Cortell-Ballester2,
Helios Pareja-Galeano1, Giuseppe Banfi3, and Giuseppe Lippi4
Abstract
Besides the use of traditional laboratory resources, the diagnosis of anemia can also be accomplished by assessing hemoglobin
(Hb) concentration with point-of-care testing (POCT) devices such as the HemoCue test systems. In several situations,
these devices might suitably replace traditional laboratory testing, including several areas of health care where a very rapid
Hb measurement might be required to make immediate therapeutic decisions. The use of these devices, however, should
fulfill some basic criteria, including economic, clinical, and regulatory issues; appropriate training of the users and knowledge
of test requirements, performance, limitations, and potential interferences; the use of venous and arterial sampling, when
possible; and a rigorous quality assessment, which should be under the responsibility of laboratory professionals. Because
of its optimal performance along with the fact that the HemoCue is probably one of the most commonly used devices
worldwide, the aim of this article is to review the literature data about the performance of this test system as compared
with laboratory reference testing estimations and according to the biological matrix.
Keywords
delivery of health care, hemoglobin concentration, anemia, automated hemocytometers, blood donors
Introduction
Besides the use of traditional laboratory resources, the
diagnosis of anemia can also be accomplished by assessing
hemoglobin (Hb) concentration with point-of-care testing
(POCT) devices such as the HemoCue (HemoCue,
Ängelholm, Sweden) test systems. In several situations,
these devices might suitably replace traditional laboratory
testing, including several areas of health care where a very
rapid Hb measurement might be required to make immediate therapeutic decisions. For instance, it could be imperative in patients who require critical venous access, especially
neonates and those undergoing chemotherapy due to the
low amount of blood required by these devices, as well as
in natural disasters or in sports medicine.
The reference method for Hb measurement is still based
in analysis on automated hematology analyzers, although
the widespread use of automated hemocytometers entails
major costs for blood tubes and reagents, as well as greater
volumes of blood and a longer turnaround time. Electronic
and portable hemoglobinometers such as HemoCue provide
fast Hb results with a high degree of quality and can be
operated using battery or mains electricity. Several
HemoCue systems, such as B-Hemoglobin, Hb 201+, 201
DM, Hb 301, and Donor Hb Checker, are available on the
market (Table 1). Hb 201+ and 201 DM employ the
azide-methemoglobin method,1 whereas Hb 301 and Donor
Hb Checker are modified versions,2 which use microcuvettes that are significantly cheaper. The measurement
range is 0 to 256 g/L, and results are obtained within 10 to
60 s. The manufacturer suggests that capillary, venous, or
arterial whole blood in ethylenediaminetetraacetic acid
(EDTA) anticoagulant can be used as the sample (www.
hemocue.com). Although many other instruments are available worldwide for mobile collection settings, including
STAT–Site MHgb (Stanbio Laboratory, Boerne, TX), Hgb
Pro Professional Hemoglobin Testing System (ITC, Edison,
NJ), D-10 Hemoglobin Testing (Bio-Rad Laboratories,
Hercules, CA), and CompoLab HB system (Fresenius Kabi
AG, Bad Homburg, Germany), scientific literature data and
1
University of Valencia, Fundación Investigación Hospital Clínico
Universitario/INCLIVA,Valencia, Spain
2
Hospital La Fe,Valencia, Spain
3
University of Milan, Milan, Italy
4
Academic Hospital of Parma, Parma, Italy
Received June 25, 2012.
Corresponding Author:
Fabian Sanchis-Gomar, Department of Physiology, Faculty of Medicine,
University of V
alencia, Av. Blasco Ibañez, 15,Valencia, 46010 Spain
Email: [email protected]
Downloaded from jla.sagepub.com by guest on November 22, 2012
2
Downloaded from jla.sagepub.com by guest on November 22, 2012
0–256
Capillary,
venous, or
arterial
Capillary,
venous, or
arterial
Capillary,
venous, or
arterial
Capillary or
venous
HemoCue Hb
201 System
HemoCue
Hb 201 DM
System
HemoCue Hb
301 System
HemoCue
Donor Hb
Checker
System
10
10
10
10
10
1000
500
Analyzer:
350; docking
station: 566
350
1000
15–40
10–40
15–30
15–30
15–30
No further
calibration
Calibration
Daily basis;
control
microcuvette
No further
calibration
Built-in “self-test” No further
calibration
Built-in “self-test” No further
calibration
Built-in “self-test” No further
calibration
Daily basis;
control
microcuvette
Operating
Temperature,
Weight, g
°C
Quality Control
ICSH, International Council for Standardization in Haematology.
105–190
0–256
0–256
0–256
Range, g/L
HemoCue
Capillary,
B-Hemoglobin venous, or
System
arterial
Sample
Sample
Volume,
µL
Table 1. Summary of Technical Characteristics of the Different HemoCue Hemoglobinometer Models
Method
Principle
Correlation of 0.99
Vanzetti Modified azidemetwhen compared with
(1966)1 hemoglobin reaction;
double-wavelength
the reference method
measuring method, 570
(ICSH method)
nm and 880 nm
Correlation of 0.99
Vanzetti Modified azidemetwhen compared with
(1966)1 hemoglobin reaction;
double-wavelength
the reference method
measuring method, 570
(ICSH method)
nm and 880 nm
Correlation of 0.99
Vanzetti Modified azidemetwhen compared with
(1966)1 hemoglobin reaction;
double-wavelength
the reference method
measuring method, 570
(ICSH method)
nm and 880 nm
Absorbance of whole
Correlation of 0.99
van
when compared with
Kampen blood at an Hb/HbO2
isobestic point; doublethe reference method and
(ICSH method)
Zijlstra wavelength measuring
(1983)2 method, 506 nm and
880 nm
Absorbance of whole
Correlation of 0.99
van
when compared with
Kampen blood at an Hb/HbO2
isobestic point; doublethe reference method and
(ICSH method)
Zijlstra wavelength measuring
(1983)2 method, 506 nm and
880 nm
Accuracy
3
Sanchis-Gomar et al.
on-field evaluation of non-HemoCue test systems are
extremely limited or even lacking for some of them. As
such, its optimal performance along with the fact that the
HemoCue is probably one of the most commonly used
devices worldwide has led us to focus this review article on
this test system.
Although the HemoCue was found to have good sensitivity and specificity for anemia,3,4 these conclusions were
reached without much consideration about the conditions
under which the test is supposed to be performed.
Accordingly, high humidity has been shown to bias
HemoCue microcuvette function as well as Hb measurements.5 Experimental evidence also attests that some discrepancies exist between capillary, venous, and arterial
sample Hb determinations. Similarly, various and even controversial results have been reported for accuracy and reliability. Therefore, the aim of the article is to review the
literature data about the performance of HemoCue devices
since we note that there are discrepancies in published literature so far, and there is evident need for summarization
of all findings.
HemoCue Experimental Evidence
In 1998, Hb values obtained using the HemoCue (the specific model was not reported) were compared with results
obtained by the Coulter Max-M (Beckman Coulter, Brea,
CA) in a clinical laboratory. Hb was measured in 52 arterial
blood samples from 13 patients during aortic surgery. No
significant differences were observed between test results
(p = 0.1). Thus, it was also found that the lower and upper
limits of agreement between the two analyzers were –0.37
and +0.45 g/dL, which led the authors to conclude that the
HemoCue provided Hb results comparable with the laboratory reference method and was hence useful for nearpatient testing.6 Another study was carried out to determine
whether HemoCue (the specific model was not specified)
may be useful in a neonatal unit to measure Hb as compared with a laboratory method (i.e., Coulter STKS;
Beckman Coulter). Samples were collected by venipuncture, by heel prick, or from arterial lines. Again, the concordance of measures between the HemoCue and laboratory
testing was excellent (limits of agreement of the two methods were between –4.8 and +9.8 g/L) over a broad range of
Hb values.7
Another study compared Hb values determined with the
HemoCue (the model was not specified) with those assessed
using laboratory instrumentation. Venous blood specimens
were collected, and good agreement was found between the
HemoCue and Technicon H3 (Bayer Technicon, Tarrytown,
NY), since the bias was within 10 g/L in 95% of measures,
and in no case did the variation exceed 20 g/L.8
Hb was also measured in venous, earstick, and fingerstick samples using the HemoCue (B-Hemoglobin) and an
automatic hematology analyzer (Abbott Cell-Dyn 3500;
Abbott Diagnostics, Abbott Park, IK). It was found that
earstick Hb results were considerably higher than the fingerstick and venous results (i.e., the mean overestimation
of the earstick hemoglobin measurement was 7.8%, with
mean bias between 7 and 28 g/L in 95% of cases), so it was
concluded that earstick collections for Hb assessment are
not advisable in routine practice. The mean Hb concentration in the fingerstick samples was instead slightly higher
than that recorded in venous samples (Pearson’s correlation coefficient r = 0.93, with bias exceeding the limit of
±10 g/L in 9% of samples). Thus, the use of fingerstick
samples led to overestimation of the actual Hb concentration, although in the authors’ opinion, this bias was still
acceptable.9
To assess the reliability of point-of-care Hb determination with HemoCue (B-Hemoglobin model) and to analyze
its usefulness for the initial diagnosis of anemia, Hb was
measured in 20 venous blood samples diluted with saline to
obtain a wide range of Hb, as well as in venous and capillary blood samples from 247 primary health care patients.
In this case, all HemoCue results were compared with those
obtained by laboratory instrumentation (Pentra 120 Retic
ABX; HORIBA Ltd., Kyoto, Japan). In diluted samples, Hb
values obtained with the HemoCue and Pentra 120 were not
significantly different (mean bias −0.1 ± 3.2 g/L) and
showed an excellent correlation (r = 0.992; p < 0.01).
HemoCue provided accurate values if at least 4 µL of blood
was loaded into the microcuvette. Moreover, no significant
differences were observed when Hb was measured in
venous and capillary blood samples. It was hence concluded
that the HemoCue may provide accurate and reliable values
in a broad range of Hb, so it can reasonably be used for the
initial screening of anemia in primary health care.10 Another
group assessed predonation venous Hb in blood donors by
capillary blood samples analyzed by the HemoCue (model
not reported) and laboratory instrumentation (Coulter MaxM; Beckman Coulter). The imprecision (i.e., mean coefficient of variation, CV) for the HemoCue was 2.3% ± 0.7%.
HemoCue showed good agreement with venous Hb (r =
0.892), although HemoCue test results exhibited a positive
bias (i.e., overestimation) of 7.8 ± 7.3 g/L (in 95% of the
cases, the results of the capillary blood sample varied from
those obtained in venous specimens, in a range between
–6.8 and 22.5 g/L), which led the authors to conclude that
capillary Hb measurement by this portable hemoglobinometer might be unreliable since it could potentially affect
both donor safety and the blood supply.11
On the other hand, the comparability of the HemoCue
(Hb 301) measurements with a Sysmex XE 2100 (Sysmex,
Kobe, Japan) was assessed by analyzing more than 300 routine venous blood samples. A bias greater than 7% occurred
in 4% of cases (i.e., 13/300), and in only 3 cases (i.e., 1%),
the bias exceeded 10%, which was considered the threshold
Downloaded from jla.sagepub.com by guest on November 22, 2012
4
Journal of Laboratory Automation XX(X)
for clinical significance by the authors. These results supported the conclusion that the HemoCue may be a suitable
strategy for blood donor screening. Additional advantages
were listed, including the lower cost of Hb 301 microcuvettes as compared with other models and the robustness
against adverse climatic conditions.9 Other authors have
compared capillary blood HemoCue test results (model not
specified) with those obtained by a Sysmex SE 9500 in
venous samples collected in patients with gastrointestinal
bleeding. The mean bias between the HemoCue and Sysmex
SE 9500 was –0.6 ± 8.7 g/L and exceeded 10 g/L in 21% of
cases, which prompted the authors to advise against the routine use of the capillary HemoCue alone for making therapeutic decisions.10 It also has been investigated whether the
HemoCue (HemoCue B-Hemoglobin) might be suitable to
assess Hb in suction fluid obtained at elective caesarean
section. The comparison of Hb values obtained in 30 women
with the HemoCue and laboratory analysis (regrettably, the
type of hemocytometer was not specified throughout the
article) exhibited a bias of –0.13 mg/L (limit of agreement
between –3.9 and 3.6 mg/L), thereby revealing a good
degree of agreement.12 In another study, the authors compared the accuracy of results obtained on fingerstick blood
samples by the HemoCue (201+ Hb model) with those
obtained on venous blood samples with a ABX Pentra 60
(HORIBA, Ltd.) in 969 unselected potential female donors.
It was found that the sensitivity of the HemoCue was only
56%, and the instrument failed to detect a relevant number
of anemic donors (up to 36%). Furthermore, the results of
capillary Hb showed a trend toward higher values than in
venous blood (overestimation of 5.9%). Finally, a poor linear correlation among ABX Pentra 60 and HemoCue was
also reported (Pearson’s correlation coefficient r = 0.716).13
A new study assessed the reliability of Hb values in
venous blood measured with the HemoCue (B-Hemoglobin)
in patients with gastrointestinal hemorrhage and compared
them with those measured on venous blood with a laboratory instrument (the model was not reported in the article).
Overall, a good correlation was observed (i.e., Pearson’s
correlation coefficient r = 0.979), and the bias was –1.0 g/L
(95% confidence interval [CI], –6.9 to 4.9 g/L). HemoCue
was hence considered a quick and reliable method for Hb
assessment in both the acute and stable phases of gastrointestinal bleeding.14 On the other hand, Richards et al.15 compared Hb values measured in venous and capillary samples
(toe and thumb) in patients undergoing caesarean section
under neuraxial anesthesia using the HemoCue (model not
reported) and laboratory instrumentation (again, the analyzer model was not reported). In this study, the mean bias
versus results obtained in venous blood samples tested in
the laboratory was –2 ± 16 g/L (HemoCue, capillary blood
from toe), –1 ± 18 g/L (HemoCue, capillary blood from
thumb), and –2 ± 16 g/L (HemoCue, venous blood).
Afterward, a new study was carried out to assess the
accuracy of HemoCue measurements (model not reported)
as compared with the Sysmex KX 21 in 535 blood donors.
The HemoCue Hb values obtained in capillary and venous
blood were also compared. The authors found that the correlation coefficient between capillary HemoCue and cell
counter values was 0.40, identical to that between capillary
and venous HemoCue. Nevertheless, the correlation coefficient between the venous HemoCue and cell counter was
0.91.16
More recently, a study compared Hb test results obtained
with the HemoCue (201+ Hb model) and Radical 7 (Masimo
Corp., Irvine, CA) in 44 patients with acute surgical hemorrhage after induction of anesthesia, during surgery according to the requirements of the anesthesiologist, and finally
after the transfer of the patient to the recovery room. A good
correlation was found between HemoCue and Radical 7
values when analyzing capillary blood (r = 0.85, p < 0.001),
with a mean bias of –1.7 ± 10.5 g/L.17 Another group compared the accuracy of the HemoCue (model not reported)
with a Beckman Coulter co-oximeter in arterial blood samples of patients who received general anesthesia for spine
surgery. In 77 of the 78 Hb measurements (i.e., 98.7%)
obtained in these patients, the HemoCue values exhibited a
bias lower than 10 g/L as compared with the co-oximeter.18
This led the authors to conclude that arterial HemoCue values might be virtually interchangeable with those obtained
with standard co-oximetry.
Accuracy and reproducibility of the HemoCue (BHemoglobin model) for Hb determination were compared
with results obtained on a Sysmex XE 2100 in children
undergoing major surgery. A total of 256 arterial blood samples were collected at several intraoperative time points.
HemoCue exhibited good reproducibility and negligible
bias when compared with the XE 2100. Potential clinically
significant differences were observed beyond a range of 20
g/L in only two cases (i.e., 0.8%). It was concluded that the
HemoCue showed reliable test results in the intraoperative
setting.19 Likewise, researchers assessed whether the noninvasive Hb measurement with the HemoCue (Hb 301) may
provide clinically acceptable accuracy in critically ill
patients when compared with a Sysmex XT 2000i. A total of
471 arterial blood samples from 65 patients were collected
and analyzed, and a capillary measurement was also performed at bedside using the same device. The mean bias
between HemoCue test results obtained on capillary blood
and those obtained on arterial blood samples with the
Sysmex XT 2000i was 2 g/L (95% CI, 1–3 g/L), and the
correlation was 0.76. Discrepancies between values greater
than 10 g/L were reported in 33% of cases. Similar data
were obtained when comparing test results obtained on
arterial blood, since the bias between HemoCue and Sysmex
XT 2000i was –1 g/L (95% CI, –0.2 to 0.2 g/L), with a correlation of 0.88 and discrepancies between values greater
than 10 g/L being reported in 31% of cases.20
Downloaded from jla.sagepub.com by guest on November 22, 2012
5
Sanchis-Gomar et al.
Hb values measured using the HemoCue (Hb 201+)
were compared with those obtained with the Beckman
Coulter LH 750. One hundred fifty blood samples were
obtained from 79 adult patients hospitalized in a surgical
intensive care unit who required an urgent Hb determination. Arterial and venous blood was analyzed using both
HemoCue and the automated laboratory analyzer, whereas
capillary blood samples were also simultaneously obtained
by fingerstick and analyzed with the HemoCue. The mean
absolute bias between the HemoCue and Beckman Coulter
LH 750 was 1 g/L (95% CI, –19 to 22 g/L) in arterial blood,
1 g/L (95% CI, –25 to 26 g/L) in venous blood, and 11 g/L
(95% CI, –36 to 58 g/L) in capillary (HemoCue) and
venous/arterial blood (Beckman Coulter LH 750). Edema
was found to be the only independent variable explaining
the discordance between capillary values obtained with the
HemoCue and those in venous or arterial blood obtained
with the hemocytometer (odds ratio 6.6; 95% CI, 2.0–22.2).
These results led the authors to advise against the use of the
HemoCue in critically patients, especially in the presence of
edema and for capillary blood.21
The accuracy of the HemoCue (Hb201+) in patients
receiving antiviral therapy after liver transplantation was
also assessed. Moreover, its usefulness in terms of cost saving and time saving was evaluated. The Hb measurements
were performed in 16 patients either in venous blood with a
Siemens ADVIA 120 (Siemens Healthcare Diagnostics,
Tarrytown, NY) or in capillary blood using the HemoCue.
Paired HemoCue measurements were also performed to
assess the imprecision of this device. Time requirements
and cost of both procedures were finally recorded and compared. The HemoCue displayed optimal reproducibility and
good correlation with the standard method (r = 0.89). The
accuracy for detecting anemia (i.e., Hb ≤100 g/L) was
excellent, as attested by the area under the receiver operator
characteristic curve (AUC, 0.96). Even more interesting,
the use of the HemoCue in this cohort of patients was associated with significant economic and time savings per
patient during follow-up.22
The suitability of the HemoCue as a POCT device for Hb
estimation in mobile blood donations and critical care areas
was also assessed. For this purpose, venous blood was collected from study participants drawn from five groups (i.e.,
preschool children, schoolchildren, pregnant women, nonpregnant women, and men) and immediately processed for
Hb measurement with the HemoCue B-Hemoglobin,
Sysmex KX21N, and the reference cyanmethemoglobin
test. The overall mean Hb was 104 g/L with the HemoCue,
103 g/L with the Sysmex KX21N, and 103 g/L with the
cyanmethemoglobin test. A very high agreement was found
for Hb test results obtained with the HemoCue and the
cyanmethemoglobin test (r = 0.995; mean bias, 1.3 g/L;
95% CI, 1.0–1.5 g/L). After adjustment for possible confounders (e.g., gender, age, and category of person), no significant difference was also observed between the HemoCue
and the cyanmethemoglobin test. A high concordance of
measures was also found between the HemoCue and
Sysmex KX21N, with a mean bias of 1.5 g/L (95% CI, –3.9
to 6.9 g/L) and a nonsignificant difference in variability
between the two measurements (p = 0.391). It was hence
suggested that the HemoCue might reliably be used as an
emergent device in critical areas as well as in those with
limited resources.23
In a recent study, the authors compared predonation capillary and venous Hb levels of 8910 first-time donors presenting for whole-blood donation. Hb testing was performed
both on fingerstick samples with the HemoCue (Donor Hb
Checker model) and on venous blood samples with the
Sysmex SE 9000. In the vast majority of blood donors, the
capillary (HemoCue) and venous (SE 9000) Hb values differed less than 10 g/L, whereas the difference exceeded 20
g/L in 86 donors (1.0%) and 30 g/L or more in 10 donors
(0.1%). Pragmatically, the categorization as having enough
or insufficient Hb for blood donation was concordant
between capillary and venous measurements in 93.3% of
females and 98.7% of males.24
The reliability of Hb measurements made with the
HemoCue (Hb 201+), compared with those made with a
Sysmex XE 2100, was assessed in critically ill patients. One
hundred ninety-eight patients were included, and a total of
1166 Hb determinations were measured using arterial blood
samples. Simultaneously, a capillary (finger or ear) measurement was performed at bedside using the HemoCue.
The HemoCue’s accuracy was not affected by the hospital
unit, puncture site (finger or ear), norepinephrine administration, or Hb levels below 80 g/L. However, the capillary
HemoCue was not sufficiently accurate to make a therapeutic decision. The method’s performance was moderately
improved by the use of arterial blood.25
To conclude, one limitation of this review is that some of
the cited publications lack a suitable statistical method
when comparing these devices.
General Considerations
There are several reasons supporting the use of POCT
devices, including those for Hb assessment, in clinical and
laboratory practice.26 First, saving time is critical in several
areas of health care, where a very rapid Hb measurement
might be required to make immediate therapeutic decisions.
This might happen in any health care context where the
clinical laboratory is too far, making turnaround time
incompatible with a fast triage (e.g., in decentralized health
care facilities with no support of a clinical laboratory unit
within a network or those organized according to a huband-spoke model), or in hospital units where the fastest
possible turnaround time from shipping a sample for Hb
assessment to the core laboratory might still be insufficient
(e.g., critical hemorrhages in the operating room, intensive
care patients). The availability of POCT for Hb assessment
Downloaded from jla.sagepub.com by guest on November 22, 2012
6
Journal of Laboratory Automation XX(X)
Table 2. Summary of Some Features of Reviewed Studies
Study
Automatic Hematology Analyzer
7
HemoCue Model
Compared
Rechner et al.
Paddle8
Gomez-Simon et al.11
Morris et al.9
Coulter Max-M
Coulter STKS
Technicon H3 analyzer
Abbott Cell-Dyn 3500
—
—
—
B-Hemoglobin
Van de Louw et al.10
Cell counter Pentra 120 Retic ABX
B-Hemoglobin
van Kampen and Zijlstra2
Coulter Max-M
Gupta et al.12
Frasca et al.20
Seguin et al.21
Marino et al.22
Sysmex XE 2100
Sysmex XE 2100
Sysmex XT 2000i
Beckman Coulter LH 750
Hb 301
B-Hemoglobin
Hb 301
Hb 201+
Nkrumah et al.23
Ziemann et al.24
Mimoz et al.25
Lippi et al.26
ADVIA 120
Sysmex KX21N
Sysmex SE 9000
Sysmex XE 2100
Hb 201+
B-Hemoglobin
Donor Hb Checker
Hb 201+
—
Accuracy and
Reproducibility
+
+
+
–
+
+
+
–
+
+
+
+
–
+/−
+/−
+
+
+/−
–
+
Type of Sample
Arterial
Capillary, venous, arterial
Venous
Capillary, venous
Capillary, venous
Capillary, venous
Venous
Arterial
Arterial
Capillary, venous, arterial
Capillary
Venous
Capillary
Capillary, arterial
The long dashes represent that the study does not provide the HemoCue model.
is also valuable due to the low amount of blood required by
these devices in patients requiring critical venous access,
especially neonates and those undergoing chemotherapy.
The use of POCT devices also represents the best option in
the unfortunate circumstance of natural disasters, where
there is a compelling need to convey laboratory technologies that can be easily transported, installed, and appropriately used outside the traditional laboratory environment.27
Sports medicine is another ideal context for POCT, where
rapid test results might guide the application of specific
training regimens28 and, even more important, might support antidoping testing when carried out with rigorous preanalytical and analytical requirements.29,30 There is, for
example, a wide experience on blood results of athletes in
comparison with those shown by blood donors based on the
HemoCue, which was proposed for evaluating possible differences between physically active and nonactive healthy
subjects.31 In all these situations, POCT devices such as the
HemoCue might yield accurate Hb results within seconds,
with a small amount of sample required and thereby less
discomfort for the athletes.
Authors’ Recommendations for
Better Accuracy
HemoCue devices have been subject to varying opinions as
to their value. Thus, for all the above-mentioned findings,
some recommendations should be taken into account:
1. Economic, clinical, and regulatory issues should
be clearly addressed before suggesting its implementation in all testing (i.e., both clinical and
laboratory) settings. Health technology assessment, usually performed in clinical circumstances
for complex and expensive drug treatments or for
imaging diagnostics techniques, should also be
applied to portable analyzers, which are widely
used and have a heavy impact on health care
quality.
2. The users of the device should be adequately
trained as with any medical device about test
requirements, performance, limitations, and
potential interferences.
3. Test results on venous and arterial sampling are
more accurate than those obtained on capillary
blood and should therefore be preferred whenever
possible (Table 2).
4. Rigorous quality assessment with either internal
quality controls and external quality assessment
or even proficiency testing should be established
for verifying and continuously monitoring the
quality of results.
5.Responsibility of assessment, quality of data,
quality control schemes, and training of personnel should be reserved to laboratory professionals,
who are experienced and trained specifically for
these purposes.
Downloaded from jla.sagepub.com by guest on November 22, 2012
7
Sanchis-Gomar et al.
Preliminary data attest that the HemoCue might be associated with favorable economic revenues. Moreover,
HemoCue devices allow quicker switch-over with good
quality and accuracy, and less time and resources are spent
on management. Thus, the HemoCue remains useful as a
clinical guide in the acute setting. More important, these
devices are leading a revolution in urban and tertiary-care
hospitals. It is, however, noteworthy that these testing
devices should not replace formal laboratory venous sampling, which still remains the gold standard.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect
to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
References
1. Vanzetti, G. An Azide-Methemoglobin Method for Hemoglobin Determination in Blood. J. Lab. Clin. Med. 1966, 67(1),
116–126.
2.van Kampen, E. J.; Zijlstra, W. G. Spectrophotometry of
Hemoglobin and Hemoglobin Derivatives. Adv. Clin. Chem.
1983, 23, 199–257.
3.Kruske, S. G.; Ruben, A. R.; Brewster, D. R. An Iron Treatment Trial in an Aboriginal Community: Improving NonAdherence. J. Paediatr. Child Health 1999, 35(2), 153–158.
4. Mills, A. F.; Meadows, N. Screening for Anaemia: Evaluation of a
Haemoglobinometer. Arch. Dis. Child. 1989, 64(10), 1468–1471.
5.Nguyen, H. T. High Humidity Affects HemoCue Cuvette
Function and HemoCue Haemoglobin Estimation in Tropical
Australia. J. Paediatr. Child Health 2002, 38(4), 427–428.
6.Lardi, A. M.; Hirst, C.; Mortimer, A. J.; McCollum, C. N.
Evaluation of the HemoCue for Measuring Intra-operative
Haemoglobin Concentrations: A Comparison with the Coulter
Max-M. Anaesthesia 1998, 53(4), 349–352.
7. Rechner, I. J.; Twigg, A.; Davies, A. F.; Imong, S. Evaluation
of the HemoCue Compared with the Coulter STKS for Measurement of Neonatal Haemoglobin. Arch. Dis. Child Fetal
Neonatal Ed. 2002, 86(3), F188–F189.
8. Paddle, J. J. Evaluation of the Haemoglobin Colour Scale and
Comparison with the HemoCue Haemoglobin Assay. Bull.
World Health Organ. 2002, 80(10), 813–816.
9.Morris, L. D.; Osei-Bimpong, A.; McKeown, D.; Roper, D.;
Lewis, S. M. Evaluation of the Utility of the HemoCue 301
Haemoglobinometer for Blood Donor Screening. Vox Sang.
2007, 93(1), 64–69.
10.Van de Louw, A.; Lasserre, N.; Drouhin, F.; Thierry, S.;
Lecuyer, L.; Caen, D.; Tenaillon, A. Reliability of HemoCue
in Patients with Gastrointestinal Bleeding. Intensive Care
Med. 2007, 33(2), 355–358.
11.Gomez-Simon, A.; Navarro-Nunez, L.; Perez-Ceballos, E.;
Lozano, M. L.; Candela, M. J.; Cascales, A.; Martinez, C.;
Corral, J.; Vicente, V.; Rivera, J. Evaluation of Four Rapid
Methods for Hemoglobin Screening of Whole Blood Donors
in Mobile Collection Settings. Transfus. Apher. Sci. 2007,
36(3), 235–242.
12. Gupta, A.; Wrench, I. J.; Feast, M. J.; Alderson, J. D. Use of
the HemoCue Near Patient Testing Device to Measure the
Concentration of Haemoglobin in Suction Fluid at Elective
Caesarean Section. Anaesthesia 2008, 63(5), 531–534.
13.Mendrone, A., Jr.; Sabino, E. C.; Sampaio, L.; Neto, C. A.;
Schreiber, G. B.; Chamone Dde, A.; Dorlhiac-Llacer, P. E.
Anemia Screening in Potential Female Blood Donors: Comparison of Two Different Quantitative Methods. Transfusion
2009, 49(4), 662–668.
14. Gomez-Escolar Viejo, L.; Sala, G. S.; Azorin, J. M.; Laudemia,
R.; Sanchez, J.; Regadera, M. P. Reliability of Hemoglobin
Measurement by HemoCue in Patients with Gastrointestinal
Bleeding. Gastroenterol. Hepatol. 2009, 32(5), 334–338.
15. Richards, N. A.; Boyce, H.; Yentis, S. M. Estimation of Blood
Haemoglobin Concentration Using the HemoCue during Caesarean Section: The Effect of Sampling Site. Int. J. Obstet.
Anesth. 2010, 19(1), 67–70.
16.Bahadur, S.; Jain, S.; Jain, M. Estimation of Hemoglobin in
Blood Donors: A Comparative Study Using HemoCue and
Cell Counter. Transfus. Apher. Sci. 2010, 43(2), 155–157.
17.Lamhaut, L.; Apriotesei, R.; Combes, X.; Lejay, M.; Carli,
P.; Vivien, B. Comparison of the Accuracy of Noninvasive
Hemoglobin Monitoring by Spectrophotometry (SpHb) and
HemoCue® with Automated Laboratory Hemoglobin Measurement. Anesthesiology 2011, 115(3), 548–554.
18.Miller, R. D.; Ward, T. A.; Shiboski, S. C.; Cohen, N. H. A
Comparison of Three Methods of Hemoglobin Monitoring
in Patients Undergoing Spine Surgery. Anesth. Analg. 2011,
112(4), 858–863.
19.Spielmann, N.; Mauch, J.; Madjdpour, C.; Schmugge, M.;
Weiss, M.; Haas, T. Accuracy and Precision of Hemoglobin
Point-of-Care Testing during Major Pediatric Surgery. Int. J.
Lab. Hematol. 2012, 34(1), 86–90.
20.Frasca, D.; Dahyot-Fizelier, C.; Catherine, K.; Levrat, Q.;
Debaene, B.; Mimoz, O. Accuracy of a Continuous Noninvasive Hemoglobin Monitor in Intensive Care Unit Patients.
Crit. Care Med. 2011, 39(10), 2277–2282.
21. Seguin, P.; Kleiber, A.; Chanavaz, C.; Morcet, J.; Malledant,
Y. Determination of Capillary Hemoglobin Levels Using the
HemoCue System in Intensive Care Patients. J. Crit. Care
2011, 26(4), 423–427.
22. Marino, Z.; Carrion, J. A.; Bedini, J. L.; Crespo, G.; Martinez,
S. M.; Sanchez-Tapias, J. M.; Forns, X.; Navasa, M. A. Evaluation of a Portable Hemoglobinometer (HemoCue) to Control
Anemia in Hepatitis C Liver Transplant Recipients Undergoing Antiviral Therapy. Eur. J. Gastroenterol. Hepatol. 2011,
23(10), 942–947.
Downloaded from jla.sagepub.com by guest on November 22, 2012
8
Journal of Laboratory Automation XX(X)
23. Nkrumah, B.; Nguah, S. B.; Sarpong, N.; Dekker, D.; Idriss,
A.; May, J.; Adu-Sarkodie, Y. Hemoglobin Estimation by the
HemoCue® Portable Hemoglobin Photometer in a Resource
Poor Setting. BMC Clin. Pathol. 2011, 11, 5.
24. Ziemann, M.; Lizardo, B.; Geusendam, G.; Schlenke, P. Reliability of Capillary Hemoglobin Screening under Routine
Conditions. Transfusion 2011, 51(12), 2714–2719.
25.Mimoz, O.; Frasca, D.; Medard, A.; Soubiron, L.; Debaene,
B.; Dahyot-Fizelier, C. Reliability of the HemoCue® Hemoglobinometer in Critically Ill Patients: A Prospective Observational Study. Minerva Anestesiol. 2011, 77(10), 979–985.
26. Lippi, G.; Plebani, M.; Favaloro, E. J.; Trenti, T. Laboratory Testing in Pharmacies. Clin. Chem. Lab. Med. 2010, 48(7), 943–953.
27.Lippi, G.; Favaloro, E. J.; Plebani, M. Laboratory Medicine
and Natural Disasters: Are We Ready for the Challenge? Clin.
Chem. Lab. Med. 2010, 48(5), 573–575.
28.Wells, H. J.; Higgins, G. L., III; Baumann, M. R. Implementing an Electronic Point-of-Care Medical Record at an
Organized Athletic Event: Challenges, Pitfalls, and Lessons
Learned. Clin. J. Sport Med. 2010, 20(5), 377–378.
29. Banfi, G.; Drago, L.; Lippi, G. Analytical Variability in Athletes Haematological Testing. Int. J. Sports Med. 2010, 31(3),
218.
30.Banfi, G.; Lombardi, G.; Colombini, A.; Lippi, G. A World
Apart: Inaccuracies of Laboratory Methodologies in Antidoping Testing. Clin. Chim. Acta 2010, 411(15–16), 1003–1008.
31. Johansson, P. I.; Ullum, H.; Jensen, K.; Secher, N. H. A Retrospective Cohort Study of Blood Hemoglobin Levels in Blood
Donors and Competitive Rowers. Scand. J. Med. Sci. Sports
2009, 19(1), 92–95.
Downloaded from jla.sagepub.com by guest on November 22, 2012