Você está na página 1de 8


VDIXXX10.1177/1040638715603087Advances in the diagnosis of bovine mastitisDuarte, Freitas, Bexiga

Review Article

Journal of Veterinary Diagnostic Investigation

Technological advances in bovine mastitis 2015, Vol. 27(6) 665­–672

© 2015 The Author(s)
Reprints and permissions:
diagnosis: an overview sagepub.com/journalsPermissions.nav
DOI: 10.1177/1040638715603087

Carla M. Duarte,1 Paulo P. Freitas, Ricardo Bexiga

Abstract. Bovine mastitis is an economic burden for dairy farmers and preventive control measures are crucial for the
sustainability of any dairy business. The identification of etiological agents is necessary in controlling the disease, reducing
risk of chronic infections and targeting antimicrobial therapy. The suitability of a detection method for routine diagnosis
depends on several factors, including specificity, sensitivity, cost, time in producing results, and suitability for large-scale
sampling of milk. This article focuses on current methodologies for identification of mastitis pathogens and for detection of
inflammation, as well as the advantages and disadvantages of different methods. Emerging technologies, such as transcriptome
and proteome analyses and nano- and microfabrication of portable devices, offer promising, sensitive methods for advanced
detection of mastitis pathogens and biomarkers of inflammation. The demand for alternative, fast, and reliable diagnostic
procedures is rising as farms become bigger. Several examples of technological and scientific advances are summarized which
have given rise to more sensitive, reliable and faster diagnostic results.

Key words: Bovine mastitis; diagnostic methods; technological advances.

Introduction detection of mastitis-associated organisms in milk, and tar-

geted use of selective media may offer significant improve-
Bovine mastitis is an economic burden for farmers because ments in sensitivity in composite cow samples and bulk tank
of decreased milk yield, premature culling, cost of veterinary milk culture.10
treatments, and other factors. Mastitis leads to changes in Phenotypic identification is based on an evaluation of
milk composition, which is dependent on the inflammatory morphology, growth characteristics, ability to metabolize
response.40 substrates, antimicrobial resistance, and other features that
The most frequent standard in measuring inflammation is result from DNA expression.81 Commercial manufacturers
cytological investigation, including milk somatic cell count of diagnostic tests have developed a number of identification
(SCC). The evaluation of milk quality for premium value or methods based on phenotypic traits, including test systemsa–d
penalties applied to milk prices is generally assessed by and other combinations of biochemical tests.81 Some advan-
SCC. Intramammary infections (IMI) are detected more fre- tages of phenotypic methods are that such methods rely on
quently through milk culturing; however, pathogen isolation biochemical characteristics that are common and associated
is not necessarily associated with inflammation. The identifi- with bacterial species, are usually easy to perform, have
cation of mastitis pathogens is of major importance in order good market availability, and have a relatively low cost. An
for adequate control measures to be taken, risk of appearance inherent weakness in phenotypic methods is that there is
of chronic infections reduced, and antimicrobial therapy tar- variability in expression of characteristics by isolates belong-
geted. This work aims to discuss the advantages and disad- ing to the same species,23 and their interpretation may be
vantages of methods used in the diagnosis of bovine mastitis subjective.9 The reproducibility of these tests is limited by
and technological advances that have the potential to offer a
“cow-side” use.
Interdisciplinary Centre of Research in Animal Health (CIISA),
Faculdade de Medicina Veterinária, Universidade de Lisboa, Lisbon,
Identification of pathogens causing Portugal (Duarte, Bexiga); Institute for Systems and Computer
intramammary infection Engineering–Microsystems and Nanotechnology (INESC–MN), Lisbon,
Portugal (Duarte, Freitas); and International Iberian Nanotechnology
Phenotypic methods. Bacterial culture has for some time Laboratory (INL), Braga, Portugal (Freitas).
served as the gold standard for the examination of pheno- 1
Corresponding Author: Carla M. Duarte, CIISA, Faculdade de
typic characteristics. Appropriate use of culture-enhance- Medicina Veterinária, Universidade de Lisboa, Av. da Universidade
ment methods can significantly increase sensitivity in the Técnica, 1300-477 Lisbon, Portugal. admargduarte@fmv.ulisboa.pt
Duarte, Freitas, Bexiga

the variability in expression and interpretation of phenotypic with SCC determination in dairy herd improvement pro-
characteristics. In addition to reproducibility, the typeability grams through the use of bronopol-preserved samples.41 A
(proportion of isolates that are assigned a type by a typing previous study74 showed, through a PCR assay, that micro-
system)62 is imperfect either at the species or at the strain biologically negative samples often contained several com-
level. Microbiological culture methods are also considered to mon mastitis pathogens, some of which displayed high
be laborious and time consuming.21 bacterial counts. Use of PCR-based tests may also be of
On-farm culture systems are increasingly used, as they interest for IMI diagnosis when culturing only detects minor
offer economic benefits to farmers by reducing therapy costs pathogens.4 Traditional PCR has advanced from detection at
and the amount of discarded milk.42 Studies on the diagnostic the reaction end-point, to detection while the reaction is
validity of on-farm culture systems showed acceptable per- occurring. This improvement was necessary because end-
formance, although the specificity recorded was relatively point collection of PCR products is not quantitative. In con-
low.52 Previous researchers42,52 compared 2 on-farm culture trast, real-time PCR is quantitative. Disadvantages of
systemse,f for clinical mastitis pathogen identification. One traditional PCR are the use of agarose gels in the detection of
systeme consisted of a culture plate with 2 different media, DNA amplification, in which resolution is very poor (~10-
one allowing the growth of both Gram-positive and Gram- fold), while real-time PCR can detect as little as a 2-fold
negative bacteria, and one selective for the growth of Gram- resolution variation.79 Another improvement in traditional
negative bacteria. The other on-farm culture systemf included PCR has been the multiplex PCR, which can simultaneously
2 media, one allowing the development of aerobic bacteria detect different genes making it potentially faster and
and another allowing the growth of coliform bacteria. Both cheaper, as all species can be amplified in a single reaction.11
methods were able to successfully categorize isolates of clin- However, multi-target analysis by PCR has 2 main constraints.
ical cases of mastitis based on their ability to differentiate First, multiplex PCR is limited in the number of targets that can
between Gram-positive and Gram-negative organisms, with be consistently amplified simultaneously because of uncontrol-
sensitivities of 97.9%e and 93.8%,f and specificities of lable primer–primer interactions. Second, identifying solution-
68.6%e and 70.1%,f respectively, but neither had the ability phase multiplex PCR amplicons typically requires a secondary
to determine if a sample was contaminated. method for the separation of size or sequence verification
Mass spectroscopy using the matrix-assisted laser desorp- prior to analysis and data interpretation,18 which may increase
tion ionization–time of flight (MALDI-TOF MS) can also be direct costs. A previous study41 asserts that the development
performed to determine bacterial species3,61 and bacterial of a PCR test capable of complementing or replacing con-
strains8 within a few minutes.53 It is a reliable, easy to use, ventional methods in IMI diagnosis presents a challenge
and cost-effective technique that has the potential to replace because of the large number of pathogens responsible for
and/or complement conventional phenotypic identification, IMI, many of which are closely related genetically. Further-
reaching sensitivity and specificity values of 100%.5 Never- more, milk contains PCR-inhibiting substances, and an assay
theless, the ability of MALDI-TOF MS in identification is designed for use in mastitic milk must include dedicated
limited to specific spectra databases of the existing bacterial DNA extraction protocols and reagents to obtain results.
protein profiles,5 and this technology is still too costly to be Molecular diagnostic methods, in general, may also help
widespread in diagnostic laboratories. identify particularly virulent strains of an organism or distin-
guish between clonal and nonclonal infection outbreaks. In a
Genotypic methods. Genotypic methods use DNA as the clonal outbreak, the predominance of a single strain could
basis for identification and are used for species identification indicate contagious transmission of the organism or expo-
and strain typing.81 The genomic sequences of a number of sure of multiple cows to a particular environmental point
mastitis-causing pathogens are now available and have been source.54 Other molecular typing methods used for bovine
used to develop nucleic acid–based testing methods, such as mastitis pathogen identification include amplified fragment
polymerase chain reaction (PCR), which has become one of length polymorphism (AFLP) analysis at a species level73;
the most popular methods for direct detection of nucleic restriction fragment length polymorphism (RFLP) analysis
acids from infectious agents.48 The high sensitivity of PCR, at a strain level64; multiple locus variable-number tandem
which is capable of detecting a single molecule of DNA, may repeat analysis (MLVA) at a strain level58; ribotyping at a
be seen as an advantage for microbiological diagnostic pur- species level12; transfer DNA intergenic spacer length poly-
poses. Up to 30% of clinical mastitis samples yield no growth morphism analysis at a species28 and strain level69; pulsed-
in bacterial culture,74 but PCR analysis is sensitive enough to field gel electrophoresis (PFGE) typing at a strain level16;
detect growth-inhibited and nonviable bacteria. This leads to and DNA sequencing of housekeeping genes at a species27
a possible decrease in the rate of false-negative results. In and strain level.28 From these 8 methods, only 3 (AFLP,73
addition, short throughput times and the potential for objec- RFLP,64 and PFGE16) have been performed directly from
tive and user-independent identification75 are other arguments milk samples but all required prior isolate recovery by micro-
in support of PCR assays. Identification of nonviable bacteria biological culture. There appears to be higher reproducibility,
has the potential to enable integration of IMI diagnostics resolution, and sensitivity to AFLP, but both this technique
Advances in the diagnosis of bovine mastitis 667

and RFLP have a similar response time and cost efficiency.55 ELISAs have been developed to screen milk for natural con-
According to a previous study,12 automated ribotyping is a tamination with S. aureus and Listeria sp. organisms.62 A
reproducible method, easy to perform, and operator-indepen- previous study80 developed a magnetic bead–based ELISA
dent. However, when performed manually, it is time-con- employing monoclonal antibodies for the detection of staph-
suming and technically demanding, requiring highly skilled ylococci in milk. This method detected between 104 and 105
personnel. The more recent typing methods provide a higher organisms/mL and took 3 hr. An earlier study51 investigated
degree of reproducibility, such as MLVA,58 triggered by the a milk antibody testg that detected S. aureus antibodies in
independent development of a large range of protocols by milk samples. The ELISA results were scored both visually
many different laboratories leading to several different typ- and by means of an optical density plate reader and com-
ing schemes for each organism. This led to interlaboratory pared against positive controls. This test had several poten-
comparisons and is one of the main limiting factors in cur- tial points of contention with microbiological tests: a cow in
rently available genotyping techniques.58 Nevertheless, the early stages of infection could be culture positive but
MLVA has a strong discriminatory capacity, high robustness, antibody negative, and a cow could be antibody positive but
portability, objectivity, and throughput34,75 but low versatil- culture negative because of the intermittent shedding pattern
ity, as most protocols are species or serotype specific. In of cows with chronic S. aureus mastitis, or because milk
comparison, PFGE, the current gold standard method for from a single infected quarter was not included (or was
molecular subtyping, has a strong discriminatory capacity diluted) in composite milk samples. If cows were ≤30 days in
and versatility, but is less robust and portable, and has lower milk or producing ≤13.6 kg of milk per day, the test was also
objectivity and throughput.34 not considered to be accurate. The sensitivity of this antibody
Other technological advances of genotypic methods for testg has been reported to range between 69% and 90%, while
the identification of bovine mastitis pathogens include specificity values were 61–97%.29 Despite this test being
microarray technology, which is capable of detecting 7 com- available for several years, its use seems to be limited.
mon species of mastitis-causing pathogens within 6 hours, Another study82 for S. aureus identification in milk
with an observed sensitivity of 94.1% and specificity of samples based on immunoagglutination compares 6 com-
100%.45 The platform used was based on PCR technology mercially available slide agglutination tests, which are
where pathogen-specific targets of DNA were amplified and currently used in human medicine. The highest sensitivity
transferred to react and hybridize with specific probes that (86.7%) and specificity (90.1%) was obtained for a test
were pre-spotted on the biochip. At the end of the process, consisting of latex particles coated with human fibrinogen
colorimetric techniques were used to identify pathogen pat- and immunoglobulin G.h
terns present in the sample. The detection limit of this method
was 103–105 CFU/mL. A previous study14 also described a
Mastitis detection
DNA chip, based on the use of a ligation detection reaction
coupled to a universal array, developed to detect and analyze Cell counting.  Somatic cell count has been used as the gold
pathogens directly from milk samples. These bacterial standard for decades to diagnose subclinical mastitis and is
groups were identified based on the 16S rRNA gene. Results an important parameter for the dairy industry as it affects the
demonstrated high specificity with sensitivity as low as 6 price of milk paid to the farmer. The mononuclear leuko-
fmol per volume unit. cytes, monocytes, and lymphocytes, along with the neutro-
Another study71 identified coagulase negative staphylo- phils, are often the only cells taken into account.57 SCCs do
cocci isolates with an updated transfer (t)DNA-PCR, which not always correlate with infection of the udder, and they
resulted in 91.0% typeability and 99.2% accuracy. The study may be affected by other factors (e.g., lactation number,
also showed that the updated tDNA-PCR associated with stage of lactation, milk production level, stress, season, and
capillary electrophoresis was almost as accurate as gene breed).66 Suggested cutoff values for SCC in mastitis diagno-
sequencing but faster (increased automation) and cheaper sis differ between publications because different conventions
(only $3 per isolate). are used in different countries as well as various types of
milk samples.32 The most frequently used cutoff value to
Immunoassays. Immunological methods are often used define subclinical mastitis is a SCC ≥200,000 cells/mL.60,66
because of their speed, simplicity, relatively low cost, and the Cell counts below this threshold in composite milk samples
availability of commercial kits.24 The detection limits of enzyme- indicate that a mammary gland is likely to be free of IMI,15
linked immunosorbent assays (ELISAs) have been shown to but this threshold is based on the assumption that the culture
range between 8 × 10−4 and 8 × 10−3 μg of antibody/mL.47 test is perfect, which does not take into account the chances
Despite these features, ELISAs are not able to detect some of false-negative results in a SCC ≤200,000 cells/mL. There-
antigens that are present at low concentrations.48 fore, a study77 determining the accuracy of both SCC and
ELISAs exist for Staphylococcus aureus detection in culture to detect IMI proposes a lower threshold of 150,000
cases of bovine mastitis,9 but the antibody titer does not cor- cells/mL, which can account for misclassification of the cul-
relate well with the amount of infecting bacteria.62 Other ture. The authors of that study suggest this is a more accurate
Duarte, Freitas, Bexiga

SCC cutoff, providing information about the prevalence of is usually used as a screening test on producer’s milk because
subclinical mastitis corresponding to test sensitivity and of its simplicity and objectivity, and also provides a conve-
specificity maximization. Even so, on an individual quarter nient method of monitoring udder health on a herd basis.
basis, a SCC cutoff point of 100,000 cells/mL may be more In 2010, an indirect method to assess SCC and fat content
appropriate57 if using differential cell counting as an alterna- in milk samples was published.19 The low-cost portable
tive method in defining the presence of mastitis. microfluidic sedimentation cytometer has a 15-min response
Somatic cell count can be measured by means of direct or time and shows a lower detection limit of 5 × 104 cells/mL.
indirect methods. Direct methods use either portable auto-
matic cell counters, which are practical for field use, or auto-
Ion variation: milk conductivity
matic counters in a laboratory setting.i There are a number of
portable cell counters available. Some counters use an ester- An effect of mastitis is changes in ion concentrations
ase-catalyzed enzymatic reactionj and othersk count somatic caused by increased vascular permeability37 leading to
cells optically by staining cell nuclei with a DNA-specific modifications in electrical conductivity of milk. Electrical
fluorescent reagent (propidium iodine). The advantages of conductivity can be measured by rising conductance in
these portable cell counters include cost-effectiveness, speed milk caused by an increase in levels of sodium, potassium,
of use, and user friendliness, but they are generally consid- calcium, magnesium, and chloride. To date, measurement
ered to have poor sensitivity at low SCC.76 The laboratory of electrical conductivity is the most widespread automated
cell counteri operates on the principle of optical fluorescence detection method for mastitis in milking robots. In such
as in the portable assayk mentioned previously, but in this systems, mastitis detection is generally performed through
case the fluorescent reagent employed is ethidium bromide. a combination of human inspection of animals, by elec-
The fluorescent signal generated is used to estimate the SCC trodes in the milking system to detect changes in milk elec-
in milk.76 When comparing both methods,i,k the automatic trical conductivity, and by data analysis in herd management
method has a higher repeatability.23 The advantage with an software to detect changes in milk yield and milking fre-
automatic cell counter is that it is objective and accurate. quency. Although milk conductivity change might be use-
Disadvantages are that it is time consuming because samples ful in detecting mastitis, it is not a reliable or sensitive
need to be sent to a laboratory, and the initial investment is parameter for conclusive diagnosis33 on its own because of
high as the equipment is very expensive. the high number of false positives.
Another direct detection method is differential cell count
(DCC), which shows changes in relative cell proportions and
Biomarker evaluation
can be used to differentiate healthy glands from inflamed
glands. DCCs are performed on quarter milk samples by A biomarker is a characteristic that can be measured and
cytometry56 and have been proposed as a valid tool for the evaluated as an indicator of normal biological processes,
identification of inflammatory processes in cases with low pathological processes, or pharmacological responses to
SCC.63 Recent studies57 have shown that DCC can reveal therapeutic interventions.6 To be considered a “good” bio-
inflammatory mastitis processes with a sensitivity and speci- marker, the indicator must be specific for a disease and
ficity of 97.3% and 92.3%, respectively, even in milk with an should remain unchanged by unrelated disorders. Likewise,
SCC of 1,000 cells/mL.56 reliable and reproducible biomarker quantification must be
The California Mastitis Test (CMT) is a common indirect demonstrated.36
method for measurement of SCC. The test is performed by As with the aforementioned ions, enzymes are also
adding a detergent to a milk sample with a high cell count, released into milk as a result of an animal’s immune response
which promotes cell lysis, nucleic acid release, and forma- against infection and changes in vascular permeability. The
tion of a “gel-like” matrix. When the cell count is above a enzymes dealing with milk synthesis tend to decrease, while
certain threshold, the sample viscosity interpretation is sub- the enzymes related to inflammation tend to increase.60 The
jective, and might result in false positives or negatives.76 A enzymes originating from phagocytes increase exponen-
sensitivity of 66.7% and specificity of 54.8% using the CMT tially. Such enzymes include N-acetyl-d-glucosaminidase
to detect IMI has been reported in fresh cows.65 The main (NAGase), β-glucuronidase, and catalase. The activity of
advantages of CMT are that it is quick, cheap, simple, and other enzymes originating in blood also increases, including
can be used as a “cow-side” test. plasminogen, which is locally activated as plasmin, a proteo-
The Wisconsin Mastitis Test (WMT) is a laboratory test lytic enzyme that degrades fibrin and casein.60 Several
generally conducted on bulk tank milk samples. The scores enzymes could be used as biomarkers for mastitis diagnosis,
can be used to predict the average number of somatic cells. including NAGase,38 serum amyloid A (SAA), haptoglobin
This indirect method uses the same reagent as the CMT; (Hp), and lactate dehydrogenase (LDH), a cytoplasmic
however, the reaction is not estimated but measured by gel enzyme.31 A previous study1 showed that LDH was the bio-
height in a tube, providing a more precise result than CMT. marker with the lowest variation between milkings in clini-
The results are generally reported in millimeters. The WMT cally healthy cows when compared with SAA, Hp, and
Advances in the diagnosis of bovine mastitis 669

NAGase. These authors assert that whatever the method On the other hand, phenotypic methods are usually con-
used, several measurements over time could be a viable sidered less expensive than genotypic methods.81 Whether or
approach as well as information on the normal biomarker not this is true, depends in part on the number of samples
variation in clinically healthy udder quarters. Colorimetric processed per time unit. In some laboratories, phenotypic
and fluorimetric assays have been developed for measuring tests are used with such high frequency that an investment in
milk enzyme concentrations, which increase during the early automation for test reading could become profitable.35
stages of mastitis, including NAGase38 and LDH.43 Regardless of sample number, additional testing may be
Milk proteins may be submitted to proteolysis caused by needed to obtain conclusive results from phenotypic methods.
bacteria or endogenous proteases during episodes of mastitis. The potential increase in cost and turnaround time of pheno-
Peptide biomarkers of milk could thus be used in the diagno- typic testing may narrow or eliminate the cost and time differ-
sis of mastitis and could discriminate between bacterial ences between phenotypic and genotypic identification.35 In
causes.49 A 2013 study49 used capillary electrophoresis, liq- addition, the cost of inaccurate results must also be taken into
uid chromatography, and mass spectrometry to reveal a bio- consideration.81 Ultimately, several genotypic methods
marker panel of 47 peptides, with a sensitivity of 75% and require investment in equipment that is usually very expen-
specificity of 100%. sive, which limits their use in routine diagnosis.
Immunoassays have also been used in the diagnosis of
bovine mastitis to detect acute phase proteins Hp and SAA,
Future trends
which increase in milk during inflammation,25 but could be
≤1 and ≤0.3 μg/mL, respectively, in healthy milk samples.2 Development prospects for new bovine mastitis diagnosis
An ELISA has been developed for Hp with a detection limit methodologies point to new biomarkers and technological
of 0.07 μg/mL in milk and serum,30 and a commercially advances for high sensitivity and specificity, fast and effi-
available solid-phase sandwich ELISA has been developed cient devices that can offer a “cow-side” use. More recently,
for SAA.72 A previous study78 used an automated optical transcriptome and proteome analyses have been introduced
biosensor–based immunoassay to detect NAGase in milk. to the biomedical research field. Such tests allow for the
The limit of detection for the assay was 1 μg/mL. Neverthe- identification of biomarkers, gene expression profiles, and
less, other researchers7 assert that while ELISAs feature the understanding of complex molecular mechanisms in cell
accuracy and specificity, antibody-based strategies are physiology and pathology.39 Advances in relevant proteomic
restricted by the ability to detect and quantify 1 protein at a techniques such as 2-dimensional gel electrophoresis and
time and by a reliance on the availability of species-specific mass spectroscopy68 have led to the identification of several
antibodies. ELISAs, therefore, have little application to the new proteins involved in mastitis. Progress in microbial pro-
discovery of novel inflammatory mediators, as currently teomics has been achieved with the availability of whole
only a limited number of bovine-specific antibodies are genome sequences for a number of bacterial groups,13 includ-
available. ing proteome profiles of mastitis-causing pathogens, which,
combined with newly available information on toxins,
enzymes, and metabolites produced in the udder, could lead
Discussion to their identification in milk.76 Proteomic studies performed
Phenotypic methods continue to be more frequently used for several mastitis pathogens have led to information on
than genotypic methods for the identification of mastitis protein expression pattern, which can be applied to the dis-
pathogens. However, we are faced with a few problems and covery of new therapeutic targets46 (e.g., bacterial immuno-
challenges. One problem is the large proportion of milk sam- genic proteins for vaccines) and new diagnostic biomarkers.
ples submitted to bacteriological analysis from mastitis cases Biosensors are fast becoming the next generation of tools
that lead to no-growth results.74 These results are generally in analyzing areas such as environmental research, medicine,
recognized as false negatives corresponding to detection fail- biodefense, agriculture, and food control.44 Biosensors use
ure of IMI causative agents.41 Therefore, a considerable biological receptor molecules (e.g., antibody, enzyme, and
number of infected cows may remain undetected without a nucleic acid) combined with a transducer to produce a signal
concomitant increase in SCC,67 and mastitis pathogen con- that shows a specific biological event (e.g., an antibody–anti-
trol and eradication in herds may be compromised.14 Conse- gen interaction). Nanotechnology-based pathogen detection
quently, phenotypic identification is being supplemented has created an array of technologies that have advanced
with genotypic methods. A 2013 study61 supports the combi- detection, diagnosis, and imaging of biomarkers of disease
nation of conventional microbiology as first identification pathogenesis,17 shortening the time span between sample
for triage and multiplex PCR use for rapid identification of uptake and results. Portability of biosensors has been
bacteria associated with IMI. Molecular detection assays are explored over the past 15 years though lab-on-chip plat-
a promising avenue in resolving false-negative issues, and forms, incorporating electronics, sampling, and detection
several assays have already been developed that can provide modules necessary for a fast, accurate, and low-cost analysis.
solutions to this problem.74 Several types of these biochips have been demonstrated,
Duarte, Freitas, Bexiga

using several detection principles: chemical,59 mechanical,20 j. Portacheck somatic cell counter, Portacheck Inc., Moorestown,
optical,26 electrical,22 and magnetic.50 NJ.
k. Cell counter, DeLaval International AB, Tumba, Sweden.

Conclusion Declaration of conflicting interests

A variety of mastitis diagnostic tests are routinely used to The author(s) declared no potential conflicts of interest with respect
evaluate microbiological milk quality in dairy farms. The to the research, authorship, and/or publication of this article.
successful choice for a test that evaluates milk requires meth-
odological knowledge and diagnostic capabilities for each Funding
test currently available. The suitability of a detection method The author(s) received no financial support for the research and
for routine diagnosis depends on its specificity, sensitivity, authorship of this article. Publication fees are funded by CIISA at
cost, amount of processing time, and suitability for a large FMV (CIISA: UID/CVT/00276/2013).
number of milk samples. New technical advances in mastitis
diagnosis still require specialized training and experience to References
interpret results. The personnel responsible should be aware
1. Åkerstedt M, et al. Natural variation in biomarkers indicating
of the strict compliance of each step in the process for good
mastitis in healthy cows. J Dairy Res 2011;78:88–96.
quality control in obtaining reliable data. 2. Åkerstedt M, et al. Relationship between haptoglobin and serum
PCR and conventional bacteriological culture are the most amyloid A in milk and milk quality. Int Dairy J 2008;18:669–674.
common tools used for pathogen detection and represent reli- 3. Barreiro JR, et al. Short communication: identification of
able diagnostic methodologies for veterinarians and farmers. subclinical cow mastitis pathogens in milk by matrix-assisted
The sensitivity of culture tests may be complemented by PCR laser desorption/ionization time-of-flight mass spectrometry.
analysis and are often combined together to yield more robust J Dairy Sci 2010;93:5661–5667.
results. However, to make treatment decisions, this combina- 4. Bexiga R, et al. Diagnosis of intramammary infection in sam-
tion does not allow for a timely answer. Proteomic research for ples yielding negative results or minor pathogens in conven-
reliable biomarkers is viable for the early detection of mastitis tional bacterial culturing. J Dairy Res 2011;78:49–55.
5. Bizzini A, Greub G. Matrix-assisted laser desorption ionization
and drug efficacy, and to discover potentially novel targets for
time-of-flight mass spectrometry, a revolution in clinical micro-
the development of alternative therapies. However, these inno-
bial identification. Clin Microbiol Infect 2010;16:1614–1619.
vations are still not possible to use for routine diagnosis. In 6. Boehmer JL. Proteomic analyses of host and pathogen
conclusion, the demands for an alternative, fast, and accurate responses during bovine mastitis. J Mammary Gland Biol
diagnostic procedure for mastitis is rising as farms increase in Neoplasia 2011;16:323–338.
size, cows produce more milk, and milking techniques such as 7. Boehmer JL, et al. The proteomic advantage: label-free quanti-
automatic milking systems become more common. fication of proteins expressed in bovine milk during experimen-
tally induced coliform mastitis. Vet Immunol Immunopathol
Authors’ contributions 2010;138:252–266.
8. Böhme K, et al. Characterization of Staphylococcus aureus
All authors contributed to conception and design of the study; con-
strains isolated from Italian dairy products by MALDI-TOF
tributed to acquisition, analysis, and interpretation of data; and
mass fingerprinting. Electrophoresis 2012;33:2355–2364.
agree to be accountable for all aspects of the work in ensuring that
9. Bourry A, Poutrel B. Bovine mastitis caused by Listeria mono-
questions relating to the accuracy or integrity of any part of the
cytogenes: kinetics of antibody responses in serum and milk
work are appropriately investigated and resolved. CM Duarte
after experimental infection. J Dairy Sci 1996;79:2189–2195.
drafted the manuscript. R Bexiga critically revised the manuscript.
10. Britten AM. The role of diagnostic microbiology in masti-
PP Freitas and R Bexiga gave final approval.
tis control programs. Vet Clin North Am Food Anim Pract
Sources and manufacturers
11. Bustin SA. Quantification of nucleic acids by PCR. In: Bustin
a. API microbial identification kits, BioMérieux Inc., Durham, SA, ed. A–Z of Quantitative PCR. (IUL Biotechnology, No. 5).
NC. La Jolla, CA: IUL Press, 2004:5–46.
b. Vitek microbial identification kits, BioMérieux Inc., Durham, 12. Carretto E, et al. Identification of coagulase-negative staphy-
NC. lococci other than Staphylococcus epidermidis by automated
c. Staph-Zym identification kit for staphylococci, Rosco ribotyping. Clin Microbiol Infect 2005;11:177–184.
Diagnostica A/S, Taastrup, Denmark. 13. Cash P. Proteomics in medical microbiology. Electrophoresis
d. BBL Crystal identification systems, BD, Franklin Lakes, NJ. 2000;21:1187–1201.
e. Minnesota Easy Culture system II bi-plate, University of 14. Cremonesi P, et al. Pathogen detection in milk samples by
Minnesota Laboratory for Udder Health, St. Paul, MN. ligation detection reaction-mediated universal array method.
f. 3M Petrifilm plate methods, 3M Microbiology, St. Paul, MN. J Dairy Sci 2009;92:3027–3039.
g. ProStaph test, Proscience Corp., Sterling, VA. 15. Dohoo IR, Leslie KE. Evaluation of changes in somatic cell
h. Masta-Staph test, Mast Group Ltd., Bootle, Merseyside, UK. counts as indicators of new intramammary infections. Prev Vet
i. Fossomatic cell counter, FOSS, Hillerød, Denmark. Med 1991;10:225–237.
Advances in the diagnosis of bovine mastitis 671

16. Douglas VL, et al. Genomic typing of Streptococcus uberis iso- 34. Hyytiä-Trees E, et al. Recent developments and future pros-
lates from cases of mastitis, in New Zealand dairy cows, using pects in subtyping of foodborne bacterial pathogens. Future
pulsed-field gel electrophoresis. Vet Microbiol 2000;75:27–41. Microbiol 2007;2:175–185.
17. Driskell JD, Tripp RA. Emerging technologies in nanotech- 35. Ieven M, et al. Rapid and economical method for species iden-
nology-based pathogen detection. Clin Microbiol Newsl tification of clinically significant coagulase-negative staphylo-
2009;31:137–144. cocci. J Clin Microbiol 1995;33:1060–1063.
18. Edwards MC, Gibbs RA. Multiplex PCR: advantages, develop- 36. Issaq HJ, Blonder J. Electrophoresis and liquid chromatog-
ment, and applications. Genome Res 1997;3:S65–S75. raphy/tandem mass spectrometry in disease biomarker dis-
19. Garcia-Cordero JL, et al. Microfluidic sedimentation cytom- covery. J Chromatogr B Analyt Technol Biomed Life Sci
eter for milk quality and bovine mastitis monitoring. Biomed 2009;877:1222–1228.
Microdevices 2010;12:1051–1059. 37. Kitchen BJ. Review of the progress of dairy science: bovine
20. Gfeller KY, et al. Micromechanical oscillators as rapid bio- mastitis: milk compositional changes and related diagnostic
sensor for the detection of active growth of Escherichia coli. tests. J Dairy Res 1981;48:167–188.
Biosens Bioelectron 2005;21:528–533. 38. Kitchen BJ, et al. Bovine milk N-acetyl-β-d-glycosaminidase
21. Gillespie BE, Oliver SP. Simultaneous detection of mastitis and its significance in the detection of abnormal udder secre-
pathogens, Staphylococcus aureus, Streptococcus uberis, and tions. J Dairy Res 1978;45:15–20.
Streptococcus agalactiae by multiplex real-time polymerase 39. Klopfleisch R, Gruber AD. Transcriptome and proteome

chain reaction. J Dairy Sci 2005;88:3510–3518. research in veterinary science: what is possible and what ques-
22. Gonçalves D, et al. Detection of DNA and proteins using amor- tions can be asked? Scientific World J 2012;2012:254962.
phous silicon ion-sensitive thin-film field effect transistors. 40. Korhonen H, Kaartinen L. Changes in the composition of milk
Biosens Bioelectron 2008;24:545–551. induced by mastitis. In: Sandholm M, et al., eds. The Bovine
23. Gonzalo C, et al. Evaluation of the overall accuracy of the Udder and Mastitis. Helsinki, Finland: University of Helsinki,
DeLaval cell counter for somatic cell counts in ovine milk. J Faculty of Veterinary Medicine, 1995:76–82.
Dairy Sci 2006;89:4613–4619. 41. Koskinen MT, et al. Analytical specificity and sensitivity of a
24. Gosling JP. A decade of development in immunoassay meth- real-time polymerase chain reaction assay for identification of
odology. Clin Chem 1990;36:1408–1427. bovine mastitis pathogens. J Dairy Sci 2009;92:952–959.
25. Grönlund U, et al. Haptoglobin and serum amyloid A in milk 42. Lago A, et al. The selective treatment of clinical mastitis based
and serum during acute and chronic experimentally induced on on-farm culture results: I. Effects on antibiotic use, milk
Staphylococcus aureus mastitis. J Dairy Res 2003;70: withholding time, and short-term clinical and bacteriological
379–386. outcomes. J Dairy Sci 2011;94:4441–4456.
26. Gunnarsson A, et al. Single-molecule detection and mis-
43. Larsen T. Determination of lactate dehydrogenase (LDH) activity
match discrimination of unlabeled DNA targets. Nano Lett in milk by a fluorometric assay. J Dairy Res 2005;72:209–216.
2008;8:183–188. 44. Lazcka O, et al. Pathogen detection: a perspective of tradi-
27. Heikens E, et al. Comparison of genotypic and phenotypic tional methods and biosensors. Biosens Bioelectron 2007;22:
methods for species-level identification of clinical iso- 1205–1217.
lates of coagulase-negative staphylococci. J Clin Microbiol 45. Lee KH, et al. Development of a novel biochip for rapid mul-
2005;43:2286–2290. tiplex detection of seven mastitis-causing pathogens in bovine
28. Heir E, et al. Identification and characterization of quaternary milk samples. J Vet Diagn Invest 2008;20:463–471.
ammonium compound resistant staphylococci from the food 46. Lippolis JD, Reinhardt TA. Utility, limitations, and promise
industry. Int J Food Microbiol 1999;48:211–219. of proteomics in animal science. Vet Immunol Immunopathol
29. Hicks CR, et al. Comparison of microbiologic culture, an
enzyme-linked immunosorbent assay, and determination of 47. Madigan MT, et al. Brock Biology of Microorganisms. 13th
somatic cell count for diagnosing Staphylococcus aureus mas- ed. San Francisco, CA: Benjamin Cummings, 2010. 1,152 p.
titis in dairy cows. J Am Vet Med Assoc 1994;204:255–260. 48. Malou N, Raoult D. Immuno-PCR: a promising ultrasensitive
30. Hiss S, et al. Haptoglobin concentrations in blood and milk diagnostic method to detect antigens and antibodies. Trends
after endotoxin challenge and quantification of mammary Hp Microbiol 2011;19:295–302.
mRNA expression. J Dairy Sci 2004;87:3778–3784. 49. Mansor R, et al. A peptidomic approach to biomarker discov-
31. Hiss S, et al. Haptoglobin and lactate dehydrogenase measure- ery for bovine mastitis. J Proteomics 2013;85:89–98.
ments in milk for the identification of subclinically diseased 50. Martins VC, et al. Challenges and trends in the development
udder quarters. Vet Med (Praha) 2007;52:245–252. of a magnetoresistive biochip portable platform. J Magn Magn
32. Holdaway RJ, et al. A comparison of indirect methods for diag- Mater 2010;322:1655–1663.
nosis of subclinical intramammary infection in lactating dairy 51. Matsushita T, et al. Performance studies of an enzyme-linked
cows. Part 2: the discriminative ability of eight parameters immunosorbent assay for detecting Staphylococcus aureus
in foremilk from individual quarters and cows. Aust J Dairy antibody in bovine milk. J Vet Diagn Invest 1990;2:163–166.
Technol 1996;51:72–78. 52. McCarron JL, et al. Laboratory evaluation of 3M Petrifilms
33. Hovinen M, et al. Accuracy and reliability of mastitis detec- and University of Minnesota Bi-plates as potential on-farm
tion with electrical conductivity and milk colour measure- tests for clinical mastitis. J Dairy Sci 2009;92:2297–2305.
ment in automatic milking. Acta Agric Scand A Anim Sci 53. Mellmann A, et al. High interlaboratory reproducibility of

2006;56:121–127. matrix-assisted laser desorption ionization-time of flight mass
Duarte, Freitas, Bexiga

spectrometry-based species identification of nonfermenting 69. Stepanović S, et al. A comparative evaluation phenotypic and molec-
bacteria. J Clin Microbiol 2009;47:3732–3734. ular methods in the identification of members of the Staphylococcus
54. Mueller UG, Wolfenbarger LL. AFLP genotyping and finger- sciuri group. Syst Appl Microbiol 2005;28:353–357.
printing. Trends Ecol Evol 1999;14:389–394. 70. Struelens MJ. Consensus guidelines for appropriate use and
55. Munoz MA, et al. Molecular epidemiology of two Klebsiella evaluation of microbial epidemiologic typing systems. Clin
pneumoniae mastitis outbreaks on a dairy farm in New York Microbiol Infect 1996;2:2–11.
State. J Clin Microbiol 2007;45:3964–3971. 71. Supré K, et al. Technical note: use of transfer RNA-intergenic
56. Pilla R, et al. Differential cell count as an alternative method to spacer PCR combined with capillary electrophoresis to identify
diagnose dairy cow mastitis. J Dairy Sci 2013;96:1653–1660. coagulase-negative Staphylococcus species originating from
57. Pilla R, et al. Microscopic differential cell counting to iden- bovine milk and teat apices. J Dairy Sci 2009;92:3204–3210.
tify inflammatory reactions in dairy cow quarter milk samples. 72. Szczubiał M, et al. Concentration of serum amyloid A and
J Dairy Sci 2012;95:4410–4420. activity of ceruloplasmin in milk from cows with clinical and
58. Pinho L, et al. Genotyping of Mycoplasma bovis isolates
subclinical mastitis. Bull Vet Inst Pulawy 52:391–395.
using multiple-locus variable-number tandem-repeat analysis. 73. Taponen S, et al. Bovine intramammary infections caused by
J Microbiol Methods 2012;88:377–385. coagulase-negative staphylococci may persist throughout lac-
59. Pinijsuwan S, et al. Sub-femtomolar electrochemical detec- tation according to amplified fragment length polymorphism-
tion of DNA hybridization based on latex/gold nanoparticle- based analysis. J Dairy Sci 2007;90:3301–3307.
assisted signal amplification. Anal Chem 2008;80:6779–6784. 74. Taponen S, et al. Real-time polymerase chain reaction-based
60. Pyörälä S. Indicators of inflammation in the diagnosis of mas- identification of bacteria in milk samples from bovine clinical
titis. Vet Res 2003;34:565–578. mastitis with no growth in conventional culturing. J Dairy Sci
61. Raemy A, et al. Phenotypic and genotypic identification of 2009;92:2610–2617.
streptococci and related bacteria isolated from bovine intrama- 75. van Belkum A, et al. Guidelines for the validation and applica-
mmary infections. Acta Vet Scand 2013;55:53. tion of typing methods for use in bacterial epidemiology. Clin
62. Riffon R, et al. Development of a rapid and sensitive test for Microbiol Infect 2007;13(Suppl 3):1–46.
identification of major pathogens in bovine mastitis by PCR. 76. Viguier C, et al. Mastitis detection: current trends and future
J Clin Microbiol 2001;39:2584–2589. perspectives. Trends Biotechnol 2009;27:486–493.
63. Rivas AL, et al. Longitudinal evaluation of bovine mammary 77. Vissio C, et al. Accuracy of the composite somatic cell count to
gland health status by somatic cell counting, flow cytometry, detect intra-mammary infection in dairy cows using latent class
and cytology. J Vet Diagn Invest 2001;13:399–407. analysis. Prev Vet Med 2014;113:547–555.
64. Saei HD, et al. Genotyping of Staphylococcus aureus isolated 78. Welbeck K, et al. Generation of an anti-NAGase single

from bovine mastitis based on PCR-RFLP analysis of the aroA chain antibody and its application in a biosensor-based assay
gene. Comp Clin Pathol 2010;19:163–168. for the detection of NAGase in milk. J Immunol Methods
65. Sargeant JM, et al. Sensitivity and specificity of somatic cell count 2011;364:14–20.
and California Mastitis Test for identifying intramammary infec- 79. Willey JM, et al. Prescott, Harley, and Klein’s Microbiology.7th
tion in early lactation. J Dairy Sci 2001;84:2018–2024. ed. New York: McGraw Hill, 2008. 1,088 p.
66. Schepers AJ, et al. Estimation of variance components for 80. Yazdankhah SP, et al. Rapid and sensitive detection of

somatic cell counts to determine thresholds for uninfected Staphylococcus species in milk by ELISA based on monodis-
quarters. J Dairy Sci 1997;80:1833–1840. perse magnetic particles. Vet Microbiol 1998;62:17–26.
67. Schwarz D, et al. Somatic cell counts and bacteriological sta- 81. Zadoks RN, Watts JL. Species identification of coagulase-neg-
tus in quarter foremilk samples of cows in Hesse, Germany—a ative staphylococci: genotyping is superior to phenotyping. Vet
longitudinal study. J Dairy Sci 2010;93:5716–5728. Microbiol 2009;134:20–28.
68. Smolenski G, et al. Characterization of host defence proteins 82. Zschöck M, et al. Evaluation of six commercial identification
in milk using a proteomic approach. J Proteome Res 2007;6: kits for the identification of Staphylococcus aureus isolated
207–215. from bovine mastitis. J Appl Microbiol 2005;98:450–455.