450014

and PlebaniJournal of Laboratory Automation

2012
JLAXXX10.1177/2211068212450014Lippi

Feature Story
Journal of Laboratory Automation
18(2) 184­–188
Continuous-Flow Automation and © 2012 Society for Laboratory
Automation and Screening

Hemolysis Index:  A Crucial Combination DOI: 10.1177/2211068212450014

jala.sagepub.com

Giuseppe Lippi1 and Mario Plebani2

Abstract
A paradigm shift has occurred in the role and organization of laboratory diagnostics over the past decades, wherein
consolidation or networking of small laboratories into larger factories and point-of-care testing have simultaneously
evolved and now seem to favorably coexist. There is now evidence, however, that the growing implementation of
continuous-flow automation, especially in closed systems, has not eased the identification of hemolyzed specimens since
the integration of preanalytical and analytical workstations would hide them from visual scrutiny, with an inherent risk
that unreliable test results may be released to the stakeholders. Along with other technical breakthroughs, the new
generation of laboratory instrumentation is increasingly equipped with systems that can systematically and automatically
be tested for a broad series of interferences, the so-called serum indices, which also include the hemolysis index.
The routine implementation of these technical tools in clinical laboratories equipped with continuous-flow automation
carries several advantages and some drawbacks that are discussed in this article.

Keywords
hemolysis, hemolyzed specimens, laboratory testing, interference

Errors in Laboratory Diagnostics
A paradigm shift has occurred in the role and organization
of laboratory diagnostics over the past decades, which has
been mainly driven by biological discoveries, incessant
implementation of novel diagnostic tests within diagnostic
reasoning, remarkable technological advances, and eco-
nomic pressures.1,2 All these forces have paved the way to
a revolution in the laboratory environment, wherein con-
solidation or networking of small laboratories into larger
factories and decentralization of some tests (i.e., point-of-
care testing) at bedside, physician offices, pharmacies,
supermarkets, and other health care facilities have simulta-
neously evolved and now seem to acceptably coexist.3,4
Contextually with these radical changes, the quality
throughout the total testing process has improved in paral-
lel, with a substantial reduction in vulnerability and error
rates. Nevertheless, laboratory diagnostics is still not the
safe enterprise it is supposed to be, and errors may occur at
an estimated frequency of ~0.3% of all tests performed.5
Although most of them are not associated with adverse
health care outcomes, in ~20% of cases, they might gener-
ate further inappropriate investigations or increase health
care expenditure, whereas in up to 6% of cases, they might
also jeopardize patient safety.6
Several lines of evidence attest that the vast majority of
mistakes occur in the extra-analytical activities of testing,
that is, those comprised within the so-called preanalytical
and postanalytical phases.7,8 The manually intensive activi-
ties of the preanalytical phase are indeed the most vulnera-
ble steps, being associated with roughly two-thirds of
all errors that can be identified throughout the total testing
process. As reliably mirrored by distribution patterns,
the vast majority of preanalytical errors occurs as a conse-
quence of mistaken or mishandled procedures while col-
lecting and handling blood specimens, which finally results
in the generation of samples of poor/unsuitable quality
for diagnostic testing.9 In terms of relative prevalence, spu-
rious hemolysis is by far the largely prevailing cause of
nonconformance (from 40%–70%), ahead of samples with
insufficient volume (from 10%–20%), those collected in
inappropriate containers (from 5%–15%) or showing undue
clotting (from 5%–12%), and those containing labeling
or identification problems (from ~2%) (Table 1).10,11
Irrespective of the large burden, hemolyzed specimens are
1
Clinical Chemistry and Hematology Laboratory, Academic Hospital of
Parma, Parma, Italy
2
Department of Laboratory Medicine, University of Padova, Padova, Italy
Received Apr 5, 2012.
Corresponding Author:
Giuseppe Lippi, Professor, U.O. Diagnostica Ematochimica, Azienda
Ospedaliero-Universitaria di Parma,Via Gramsci, 14, 43126–Parma, Italy.
Email: glippi@ao.pr.it, ulippi@tin.it

Downloaded from jla.sagepub.com by guest on February 16, 2016

Lippi and Plebani 185
Table 1. Leading Preanalytical Problems and Available Technology to Address Them.
Preanalytical Problem Prevalence, %
Hemolyzed samples 40–70
Insufficient samples 10–20
Inadequate container 5–15
Undue clotting 5–12
Misidentification ~2
Lipemic or icteric samples ~1–2
also associated with relevant clinical, organizational, col-
lective, and economical issues. The presence of cell-free
hemoglobin in the sample attributable to spurious break-
down of red blood cells while drawing blood or handling
the tube causes a kaleidoscope of biological, chemical, and
spectrophotometric interferences that globally jeopardize
the reliability of test results, especially hemostasis testing,
potassium, aminotransferases, lactate dehydrogenase, and
cardiospecific troponins.12–14 When the concentration of
cell-free hemoglobin in the sample exceeds manufacturer-,
instrument-, and analyte-specific thresholds, test results
should be suppressed, and a second—it is hoped suitable—
sample should be recollected. This obviously results in
increased costs due to the need for drawing a second blood
tube, delay in reporting results to the referring clinicians,
and frequent complaints from nurses and other health care
personnel.15 On the other hand, when left unidentified or
unmanaged, hemolyzed samples are indeed a latent hazard
that can seriously affect patient safety, so a standardized
procedure for identification and management is needed,
indeed.16
Hemolyzed Samples and
Laboratory Automation
As we move forward toward novel models of laboratory
automation and integration, new challenges emerge. The
complexity and the radical changes that have occurred in
laboratory organization have often culminated in the inte-
gration and/or consolidation of analytical and preanalytical
instrumentation. In particular, “closed systems” are gener-
ally referred to as packages of preanalytical automation
lines, analyzers, and postanalytical instruments that are
more expensive than open systems, can hardly be modified
after installation, and are characterized by poor flexibility.
The unquestionable benefits of this type of automation
can be mostly found in complete replacement of manual,
potentially dangerous, error-prone steps with automated
activities that require negligible contribution from labora-
tory staff, increased throughput, and reduced turnaround
time (TAT). There is clear evidence, however, that the
adoption of these solutions has not eased the identification
of some preanalytical problems.17 The implementation of
continuous-flow automation, where sample processing and
Downloaded from jla.sagepub.com by guest on February 16, 2016
Technology
Hemolysis index
Liquid-level sensors
Optic detection of tube or color of the stopper
Liquid-level sensors
Barcode, radiofrequency identification, delta check on results
Serum indices
analysis develop in a continuous and logical sequence of
activities (i.e., check-in, centrifugation, decapping, aliquot-
ing, analysis and storage), has in fact partially—if not
totally—hidden the samples from visual scrutiny of medi-
cal technologists and laboratory technicians, with the inher-
ent risk that unsuitable samples are left unidentified and
thereby loaded and processed. Considering that phleboto-
mists and nurses have a varying perception on whether a
sample is suitable for testing at the time of blood with-
drawal (e.g., Stauss et al18 recently reported that it is more
likely that a sample is unsuitable when the phlebotomist
thinks it is suitable and vice versa), innovative technical
tools should be put forward and integrated within labora-
tory automation to replace visual scrutiny before blood
specimens are analyzed. A variety of technological advances
have recently improved the automatic detection of the most
common sources of nonconformity. One of the former solu-
tions has been the introduction of liquid-level sensors in the
instruments’ needle to limit aspiration errors in samples
with low to insufficient volume or detect the presence of
microclots or bubbles. Another important tool is the imple-
mentation of barcode or even RFID (radiofrequency identi-
fication) technology, which now allows a bidirectional
communication between instrumentation and laboratory
information system (LIS), to prevent identification and
transcriptional errors (Table 1).
Regardless of these breakthroughs, reliable detection
and management of unsuitable specimens remain the major
tasks in laboratory diagnostics, and their identification,
especially of those plagued by spurious hemolysis, remains
rather challenging in “closed” automation systems (Figure
1). The recognition of hemolyzed samples has been tradi-
tionally accomplished by visual inspection and by direct
comparison with standardized color-coded scales to deter-
mine the degree of hemolysis and thereby roughly estimate
the content of cell-free hemoglobin. Although largely per-
fectible due to the high arbitrariness of judgment,19,20 this
technique has nevertheless represented the “gold standard”
for decades. Together with the technical innovations previ-
ously discussed, the new generation of analytical and pre-
analytical workstations is increasingly equipped with
systems that can systematically and automatically test for a
broad series of interferences, the so-called serum indices,
which also include the hemolysis index (HI).21
186 Journal of Laboratory Automation 18(2)

Figure 1. Identification of hemolyzed specimens in “closed” continuous-flow automation.

Some differences exist in the way the HI is assessed and
reported, but most instruments can provide a qualitative or
quantitative measure of cell-free hemoglobin in the sample,
which can be further compared with specific thresholds that
would help determine if the sample is suitable or not for
testing. Although the assessment of this type of interference
by using multiple wavelengths may be preferable,22 this
approach is unavailable on most automated analyzers, so
measurement of the HI is usually based on monitoring of
serum or plasma absorbance at a bichromatic wavelength,
most commonly between 570 and 600 nm (the serum indi-
ces for lipemia and icterus are instead typically monitored
at bichromatic wavelengths between 660–700 and 480–505
nm, respectively).23 By solving a set of predefined equa-
tions, the HI is finally calculated and expressed as either a
quantitative value (e.g., cell-free hemoglobin in arbitrary
units or g/L) or a qualitative index (e.g., from “0” to “+”).
The medical technologist can also customize the cutoffs for
generating specific alerts. A recent evaluation of the HI on
seven different automated clinical chemistry instruments
has documented excellent performances in terms of impre-
cision (i.e., intra-assay coefficient of variation between
0.1% and 2.7%) and interlaboratory agreement.24 Specific
external quality assessment (EQA) schemes based on ship-
ment of frozen materials can also be settled to harmonize
and monitor interlaboratory performance.25
Advantages and Limitations of the
Hemolysis Index
There are some obvious advantages in the widespread
adoption of the HI in routine laboratory practice (Table 2).
These include (a) overcoming visual inspection and arbi-
trary judgment about sample quality; (b) transmission of
the HI to the LIS, where the data are safely archived for
both forensic reasons (i.e., complaints from the wards) and
practical reasons (i.e., potential inclusion in the laboratory
report); and (c) assessment of sample quality in high-
volume clinical laboratories that use continuous-flow auto-
mation with integration of preanalytical and analytical
workstations, where automatic/systematic monitoring of
serum indices represents the only suitable approach to
detect unsuitable specimens in “real time,” before they are
released from conveyor belts. Another important advantage
of the HI is that it can be used as a surrogate to judge trans-
port time, transport temperature and centrifugation effi-
ciency, and assess phlebotomists’ performance, wherein
individual phlebotomists or phlebotomy centers encounter-
ing a high burden of spuriously hemolyzed specimens can
be objectively identified and further subjected to technical
(e.g., replacement of the material used for drawing blood)
or educational (e.g., specific training) interventions.26 The
routine use of the HI, however, has some drawbacks. First,

Downloaded from jla.sagepub.com by guest on February 16, 2016

Lippi and Plebani 187
Table 2. Advantages and Potential Limitations of Routine
Implementation of the Hemolysis Index in Clinical Laboratories
Designed around Continuous-Flow Automation.
Advantages
• Overcoming visual inspection and arbitrary judgment about
sample quality
•A  utomatic transmission of hemolysis degree to the
laboratory information system (LIS)
•A  ssessment of sample quality in high-volume clinical
laboratories where preanalytical and analytical workstations
are integrated
• S urrogate measure to judge phlebotomists’ performance
Limitations
• Unclear impact on the turnaround time
• Not validated for diagnostic purposes
• Should not be used for adjusting test results
we know of no study that has established how the system-
atic measurement of HI on all samples would affect TAT.
This may be relevant in terms of throughput and efficiency
since the HI is performed as an add-on test in some auto-
mated instrumentations. Moreover, although the HI is often
provided by manufacturers also for immunoassays, it is
typically part of the chemistry (colorimetric) analyzer,
which would hence lead to inefficiencies when the degree
of hemolysis in samples received only for immunoassay
has to be checked with a necessary transit of the specimens
through the chemistry module(s). It is then noteworthy that
this measure has not been validated for diagnostic purposes
(e.g., for diagnosing and monitoring hemolytic anemia or
for assessing blood substitutes), and it can only be used for
obtaining information on sample quality. Although specific
software might be available for automatic correction of test
results carried out on hemolyzed specimens according to
the degree of hemolysis, and some laboratories have actu-
ally incorporated the HI into their reporting algorithms, the
HI must not be used for adjusting test results. We have
recently proven that the mechanical injury of erythrocytes
and other blood cells that may arise during collection, han-
dling, transportation, or storage of blood specimens does
not occur homogeneously (i.e., it is largely dependent on
the individual erythrocyte fragility),27 nor is the intracellu-
lar content of several molecules (especially ions, enzymes,
hemoglobin, and other proteins) identical and hence suffi-
ciently predictable among different subjects, which would
make the use of corrective formulas based on the HI of the
sample unreliable and even misleading.28 Finally, manufac-
turers often provide hemolysis limits for their assays and
assume that the laboratory will take notice of the HI of that
assay. It is therefore the laboratories’ duty to comply with
that stated use.
Most clinical laboratories, especially those with the larg-
est volume of samples, clearly profit from total laboratory
Downloaded from jla.sagepub.com by guest on February 16, 2016
automation and integrated preanalytical-analytical systems
because of the minor handling of specimens, prevention of
mislabeling,29 and possibility of relieving medical technol-
ogists and medical laboratory technicians of repetitive
activities such as transporting specimens and loading ana-
lyzers.30 Besides detailed analysis of workflow to make
laboratory automation as efficient as possible, the process
also requires a straightforward knowledge of the potential
sources of poor-quality specimens as well as the best solu-
tions to identify and manage this foremost preanalytical
problem. Continuous-flow automation indeed represents a
remarkable improvement for laboratory organization, which
requires systematic monitoring of sample quality. Special
consideration hence should be given to routine implementa-
tion of serum indices in clinical laboratories equipped with
continuous-flow automation.
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, author-
ship, and/or publication of this article.
References
1. Zaninotto, M.; Plebani, M. The “Hospital Central Labora-
tory”: Automation, Integration and Clinical Usefulness. Clin.
Chem. Lab. Med. 2010, 48, 911–917.
2. Brunetti, M.; Pregno, S.; Schünemann, H.; Plebani, M.; Trenti,
T. Economic Evidence in Decision-Making Process in Labo-
ratory Medicine. Clin. Chem. Lab. Med. 2011, 49, 617–621.
3. Lippi, G.; Plebani, M.; Favaloro, E. J.; Trenti, T. Laboratory Test-
ing in Pharmacies. Clin. Chem. Lab. Med. 2010, 48, 943–953.
4. Plebani, M.; Lippi, G. Is Laboratory Medicine a Dying Profes-
sion? Blessed Are Those Who Have Not Seen and Yet Have
Believed. Clin. Biochem. 2010, 43, 939–941.
5. Carraro, P.; Plebani, M. Errors in a Stat Laboratory: Types
and Frequencies 10 Years Later. Clin. Chem. 2007, 53, 1338–
1342.
6. Plebani, M.; Carraro, P. Mistakes in a Stat Laboratory: Types
and Frequency. Clin. Chem. 1997, 43, 1348–1351.
7. Plebani, M.; Lippi, G. Closing the Brain-to-Brain Loop in
Laboratory Testing. Clin. Chem. Lab. Med. 2011, 49, 1131–
1133.
8. Plebani, M.; Lippi, G. Improving the Post-Analytical Phase.
Clin. Chem. Lab. Med. 2010, 48, 435–436.
9. Lippi, G.; Chance, J. J.; Church, S.; Dazzi, P.; Fontana, R.;
Giavarina, D.; Grankvist, K.; Huisman, W.; Kouri, T.; Palicka,
V.; et al. Preanalytical Quality Improvement: From Dream to
Reality. Clin. Chem. Lab. Med. 2011, 49, 1113–1126.
10. Lippi, G.; Guidi, G. C.; Mattiuzzi, C.; Plebani, M. Preanalyti-
cal Variability: The Dark Side of the Moon in Laboratory Test-
ing. Clin. Chem. Lab. Med. 2006, 44, 358–365.
188 Journal of Laboratory Automation 18(2)
11. Lippi, G.; Guidi, G. C. Risk Management in the Preanalytical
Phase of Laboratory Testing. Clin. Chem. Lab. Med. 2007, 45,
720–727.
12. Lippi, G.; Blanckaert, N.; Bonini, P.; Green, S.; Kitchen, S.;
Palicka, V.; Vassault, A. J.; Plebani, M. Haemolysis: An Over-
view of the Leading Cause of Unsuitable Specimens in Clini-
cal Laboratories. Clin. Chem. Lab. Med. 2008, 46, 764–772.
13. Lippi, G.; Fontana, R.; Avanzini, P.; Aloe, R.; Ippolito, L.;
Sandei, F.; Favaloro, E. J. Influence of Mechanical Trauma
of Blood and Hemolysis on PFA-100 Testing. Blood Coagul.
Fibrinolysis 2012, 23, 82–86.
14. Lippi, G.; Avanzini, P.; Dipalo, M.; Aloe, R.; Cervellin, G. Influ-
ence of Hemolysis on Troponin Testing: Studies on Beckman
Coulter UniCel Dxl 800 Accu-TnI and Overview of the Litera-
ture. Clin. Chem. Lab. Med. 2011 Sep 6. [Epub ahead of print]
15. Lippi, G.; Plebani, M.; Di Somma, S.; Cervellin, G. Hemo-
lyzed Specimens: A Major Challenge for Emergency Depart-
ments and Clinical Laboratories. Crit. Rev. Clin. Lab. Sci.
2011, 48, 143–153.
16. Simundic, A. M.; Topic, E.; Nikolac, N.; Lippi, G. Hemolysis
Detection and Management of Hemolysed Specimens. Bio-
chem. Med (Zagreb). 2010, 20, 154–159.
17. Lam, C.W.; Jacob, E. Implementing a Laboratory Automation
System: Experience of a Large Clinical Laboratory. J. Lab.
Autom. 2012, 17, 16–23.
18. Stauss, M.; Sherman, B.; Pugh, L.; Parone, D.; Looby-
Rodriguez, K.; Bell, A.; Reed, C. R. Hemolysis of Coagula-
tion Specimens: A Comparative Study of Intravenous Draw
Methods. J. Emerg. Nurs. 2012, 38, 15–21.
19. Glick, M. R.; Ryder, K. W.; Glick, S. J.; Woods, J. R. Unreli-
able Visual Estimation of the Incidence and Amount of Tur-
bidity, Hemolysis, and Icterus in Serum from Hospitalized
Patients. Clin. Chem. 1989, 35, 837–839.
20. Simundic, A. M.; Nikolac, N.; Ivankovic, V.; Ferenec-Ruzic,
D.; Magdic, B.; Kvaternik, M.; Topic, E. Comparison of Visual
vs. Automated Detection of Lipemic, Icteric and Hemolyzed
Specimens: Can We Rely on a Human Eye? Clin. Chem. Lab.
Med. 2009, 47, 1361–1365.
Downloaded from jla.sagepub.com by guest on February 16, 2016
21. Plebani, M.; Lippi, G. Hemolysis Index: Quality Indicator or
Criterion for Sample Rejection? Clin. Chem. Lab. Med. 2009,
47, 899–902.
22. Harboe, M. A Method for Determination of Hemoglobin in
Plasma by Near-Ultraviolet Spectrophotometry. Scand. J.
Clin. Lab. Invest. 1959, 11, 66–70.
23. Golf, S. W.; Schneider, S.; Friemann, E.; Temme, H.;
Roka, L. Correction of Catalytic Activities of Aspartate-
Aminotransferase, Lactate-Dehydrogenase, Acid-Phosphatase
and Potassium Concentration in Hemolytic Plasma by Deter­
mination of Hemoglobin Concentration with Direct Spectro-
photometry. J. Clin. Chem. Clin. Biochem. 1985, 23, 585.
24. Lippi, G.; Luca Salvagno, G.; Blanckaert, N.; Giavarina, D.;
Green, S.; Kitchen, S., Palicka, V.; Vassault, A. J.; Plebani, M.
Multicenter Evaluation of the Hemolysis Index in Automated
Clinical Chemistry Systems. Clin. Chem. Lab. Med. 2009, 47,
934–939.
25. Miller, W. G.; Erek, A.; Cunningham, T. D.; Oladipo, O.;
Scott, M. G.; Johnson, R. E. Commutability Limitations Influ-
ence Quality Control Results with Different Reagent Lots.
Clin. Chem. 2011, 57, 76–83.
26. Lippi, G.; Plebani, M. Reduction of Unsuitable Specimens: A
More Radical and Comprehensive Approach Is Needed. Clin.
Chim. Acta 2011, 412, 400.
27. Lippi, G.; Mercadanti, M.; Aloe, R.; Targher, G. Erythrocyte
Mechanical Fragility Is Increased in Patients with Type 2 Dia-
betes. Eur. J. Intern. Med. 2012, 23, 150–153.
28. Lippi, G.; Avanzini, P.; Pavesi, F.; Bardi, M.; Ippolito, L.; Aloe,
R.; Favaloro, E. J. Studies on In Vitro Hemolysis and Utility
of Corrective Formulas for Reporting Results on Hemolyzed
Specimens. Biochem. Med (Zagreb). 2011, 21, 297–305.
29. Lippi, G.; Sonntag, O.; Plebani, M. Appropriate Labelling
of Blood Collection Tubes: A Step Ahead towards Patient’s
Safety. Clin. Chem. Lab. Med. 2011 Sep 30. [Epub ahead of
print]
30. Plebani, M.; Lippi, G. Darwinian Evolution or Regression?
The Fate of Laboratory Professionals. Clin. Chem. Lab. Med.
2010, 48, 1367–1368.