Bacteriological etiology and treatment of mastitis in Finnish dairy herds

Vakkamäki e al. Ac a Ve Scand (2017) 59:33 DOI 10.1186/s13028-017-0301-4 RESEARCH Bac e iological e iology and ea men o mas i is inFinnish dai y he ds Johanna Vakkamäki1*, Su i Taponen1, Anna‑Maija Heikkilä2 and Sa u Pyö älä1 Abs ac Backg ound: The Finnish dai y he d eco ding sys em main ains p oduc ion and heal h eco ds o cows and he ds. Ve e ina ians and a me s egis e e e ina y ea men s in he sys em. Milk samples o mic obiological analysis a e ou inely aken om mas i ic cows. The labo a o y o he la ges dai y company in Finland, Valio L d., analyzes mos samples using eal‑ ime PCR. This s udy add essed pa hogen‑speci ic mic obiological da a and ea men and cull‑ ing eco ds, in combina ion wi h cow and he d cha ac e is ics, om he Finnish dai y he d eco ding sys em du ing 2010–2012. Resul s: The da a de i ed om 240,067 qua e milk samples om 93,529 dai y cows wi h mas i is; 238,235 cows om he same he ds se ed as he con ol g oup. No a ge pa hogen DNA was de ec ed in 12% o he samples. In 49% o he posi i e samples, only one a ge species and in 19%, wo species wi h one dominan species we e p esen . The mos common species in he samples wi h a single species only we e coagulase‑nega i e s aphylo‑ cocci (CNS) (43%), ollowed by S aphylococcus au eus (21%), S ep ococcus ube is (9%), S ep ococcus dysgalac iae (8%), Co ynebac e ium bo is (7%), and Esche ichia coli (5%). On a e age, 36% o he s udy cows and 6% o he con ol cows had eco ded mas i is ea men s du ing lac a ion. The co esponding p opo ions we e 16 and 6% a d ying‑o . Fo mo e han 75% o he ea men s du ing lac a ion, diagnosis was acu e clinical mas i is. In he milk samples om cows wi h a eco ded mas i is ea men du ing lac a ion, CNS and S. au eus we e mos common, ollowed by s ep ococci. Al oge he , 48% o he cows we e culled du ing he s udy. Mas i is was epo ed as he mos common eason o cull; 49% o s udy cows and 18% o con ol cows we e culled because o mas i is. Culling was mos likely i S. au eus was de ec ed in he milk sample submi ed du ing he culling yea . Conclusions: The PCR es has p o en o be an applicable me hod also o la ge‑scale use in bac e ial diagnos ics. In he p esen s udy, mic obiological diagnosis was unequi ocal in he g ea majo i y o samples whe e a single species o wo species wi h one domina ing we e de ec ed. Coagulase‑nega i e s aphylococci and S. au eus we e he mos common species. S. au eus was also he mos common pa hogen among he culled cows, which emphasizes he impo ance o p e en i e measu es. Keywo ds: Bo ine mas i is, Bac e iology, PCR, T ea men , Culling © The Au ho (s) 2017. This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/ publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. Backg ound Du ing ecen decades, conside able s uc u al changes ha e occu ed in he dai y indus y in many coun ies. In Finland, he o al numbe o dai y a ms has dec eased subs an ially bu a e age he d size and numbe s o a ms wi h au oma ic milking sys ems (AMS) ha e inc eased [1–3]. Wi h inc easing he d size, good managemen p ac ices and ca e o he cows become e e mo e impo - an [4, 5]. Despi e decades o mas i is con ol p og ams in coun ies wi h an in ensi e dai y indus y, mas i is con inues o be he mos common and economically impo an disease o dai y cows. I conside ably a ec s milk p oduc ion, animal wel a e, and ood sa e y [6–8]. Today’s consume s make inc easing demands o ade- qua e animal wel a e as well as ood quali y and sa e y, which also mus be aken in o accoun on mode n dai y a ms [9]. Open Access Ac a Ve e ina ia Scandina ica *Co espondence: johanna. akkamak[email p o ec ed] 1 Depa men o P oduc ion Animal Medicine, Facul y o Ve e ina y Medicine, Uni e si y o Helsinki, Pa oninkuja 20, 04920 Saa en aus, Finland Full lis o au ho in o ma ion is a ailable a he end o he a icle Page 2 o 9 Vakkamäki e al. Ac a Ve Scand (2017) 59:33 Mas i is esul s om in amamma y in ec ion (IMI), mos ly caused by a ious bac e ial species. Knowledge on he bac e iological e iology o IMI is impo an o e icien mas i is con ol and ea men [10, 11]. Dis i- bu ion o he pa hogens causing IMI a ies among coun- ies and he ds. The mos common causes o IMI a e he majo pa hogens Esche ichia coli, S aphylococcus au eus, and en i onmen al s ep ococci, and he mino pa hogens coagulase-nega i e s aphylococci (CNS) [12–15]. P e i- ous s udies on he mic obiological e iology o mas i is o IMI we e mainly based on sampling selec ed he ds in di e en coun ies. In Finland, na ional su eys wi h a andom sample om all Finnish dai y he ds we e pub- lished [16]. In a p e ious Finnish s udy, Koi ula e al. [17] used a smalle , cul u e-based bac e iological da abase o simila ype as in his s udy. Some s udies in es iga ed he ela ionship be ween he d ac o s, such as gene al managemen , milking echnique o p oduc ion ype, and dis ibu ions o mas i is-causing bac e ial species in he he ds [15, 18–20]. To ou knowledge, s udies combining pa hogen-speci ic bac e iological mas i is da a, ea men and culling eco ds, and cow and he d cha ac e is ics om la ge na ional da abases ha e no been ca ied ou . The Finnish dai y he d eco ding sys em has long adi- ions: i began in 1898 [21]. In 2015, he eco ding sys em co e ed 74% o dai y he ds and 81% o dai y cows [22]. In he ds belonging o his sys em, p oduc ion eco ds, including milk yield, milk p o ein con en , milk a con- en and milk soma ic cell coun (SCC), a e eco ded o e e y cow. The na ional heal h moni o ing o dai y he ds, cons i u ed du ing he ea ly 1980s, keeps heal h eco ds o e e y Finnish cow. Ve e ina ians and a me s egis e all e e ina y ea men s pe cow o he d in his sys em. The No dic dai y he d heal h eco ding sys em is glob- ally unique. In many coun ies, mas i is ea men s a e no eco ded o a e eco ded a he d le el only, o in con- nec ion wi h a speci ic heal h p ojec . Un o una ely, he comple eness o he eco ds o clinical mas i is in Finland, acco ding o a s udy ca ied ou in 2008, was only 56%, which was lowe han in he o he No dic coun ies [23]. In Finland, milk samples o bac e iology a e ou inely aken in mos cases o mas i is [24]. The labo a o y o he la ges dai y company in Finland, Valio L d., analyzes he majo i y o hese samples: 78% o Finnish dai y p oduce s send milk samples o his labo a o y (Pe sonal communi- ca ion K is iina Sa joka i, Valio L d., 2016). Since 2010, a mul iplex eal- ime polyme ase chain eac ion (PCR) assay has been used o he mic obiological analyses. Ou i s aim was o use a la ge mic obiological da a- base o qua e milk samples om mas i ic dai y cows o s udy he bac e iological e iology o mas i is in Fin- land. Ou second aim was o s udy ea men and cull- ing eco ds o he cows in he s udy he ds collec ed om he da abases o he Finnish dai y he d eco ding sys em. De ailed analysis o he ela ionships be ween pa hogen- speci ic IMI and di e en cow- and he d-speci ic ac o s a e p esen ed in Taponen e al. [25]. The p esen s udy p o ides in o ma ion abou a PCR assay used o ou ine mic obiological diagnos ics o mas i ic milk samples. Me hods Mic obiological da a Mic obiological diagnoses o qua e milk samples om mas i ic cows analyzed a he labo a o y o Valio L d. du - ing 2010, 2011 and 2012 we e ecei ed om he da abase o Valio L d. Fo mic obiological analyses, Pa hoP oo ™ Mas i is PCR Comple e-12 assay (The mo Fishe Scien- i ic, Wal ham, MA, USA) was used, which con ained oligonucleo ides o he s aphylococcal β-lac amase gene (blaZ) and o he ollowing mic obial species o g oups o species: Co ynebac e ium bo is, En e ococ- cus spp., including En e ococcus aecalis and En e ococ- cus aecium, E. coli, Klebsiella oxy oca and Klebsiella pneumoniae, Se a ia ma cescens, S aphylococcus spp., including all ele an CNS, S. au eus, S ep ococcus aga- lac iae, S ep ococcus dysgalac iae, S ep ococcus ube is, and T uepe ella pyogenes and Pep oniphilus indolicus. In Ma ch 2012, he es ki was changed o Pa hoP oo ™ Mas i is PCR Comple e-16 assay, which also includes Mycoplasma (M.) spp., Mycobac e ium bo is, P o o heca spp., and yeas s. The eason o milk sampling was e idence o clinical o subclinical mas i is (ele a ed milk SCC) in he qua - e . Milk sample was aken by he d s a o a isi ing e - e ina ian using an asep ic echnique [26]. In his s udy, a cow wi h a qua e milk sample wi h DNA o a leas one a ge pa hogen species in he PCR assay was consid- e ed o ha e IMI. These cows comp ise he s udy g oup. Only samples wi h mic obial DNA o one mic obial spe- cies alone we e included in calcula ion o he p opo ions o di e en pa hogens. The con ol g oup comp ised all o he cows om he same he ds: hose o which no milk samples we e sen o mic obiological analysis o he lab- o a o y o Valio L d. Bac e iological da a o 2010–2012 consis ed o mic o- biological esul s om 240,067 qua e milk samples (Table1). Samples we e aken om 93,529 dai y cows in 4725 dai y he ds. The con ol g oup consis ed o 238,235 cows om he same dai y he ds. In he analyses ca ied ou on an annual basis, he same cow may appea se e al imes. Milk andheal h eco ding da a Milk and heal h eco ding da a o he cows we e ecei ed om he da abases o he Finnish dai y he d eco ding sys em, and he Finnish ca le heal h moni o ing sys em. Page 3 o 9 Vakkamäki e al. Ac a Ve Scand (2017) 59:33 The da a include in o ma ion on indi idual cows o b eed, age, pa i y, annual milk, a and p o ein yield, esul s o es milking six o wen y- ou imes a yea , he numbe o eeding days, da e o diagnosis and ea - men o mas i is (clinical, subclinical) o d y cow he - apy (DCT), and da e and eason o culling. The da a also p o ide he d-speci ic in o ma ion, such as ha o milk- ing and housing sys em and p oduc ion ype (o ganic s. con en ional) o 93% o he cows in he da ase . In Finland and o he No dic coun ies, only e e ina - ians a e pe mi ed o p esc ibe an imic obials o use in animals. Acco ding o Finnish legisla ion, e e ina ians eco d ea men o a cow a e adminis e ing p oduc s wi h a wi hd awal pe iod, such as an imic obials. T ea - men eco ds a e egula ly submi ed o he cen al ca - le heal h da abase. In subclinical o mild clinical mas i is, he d s a o en ake a milk sample and send i o mic obi- ological analysis and con ac hei e e ina ian a e ecei - ing he mic obiological esul . Mas i is ea men in such cases may be p esc ibed wi hou he e e ina ian isi ing he a m, and he he d s a should eco d he ea men o ha cow. Ve e ina y-supe ised cases a e eco ded using ou di e en codes o mas i is du ing lac a ion: acu e clinical, subclinical, ch onic, and unspeci ied mas i is, and one code o DCT [27]. He d s a use h ee addi ional codes o eco d ea men o mas i is: mas i is du ing lac- a ion, DCT o mas i is, and DCT o p e en ion. Finnish legisla ion conce ning an imic obial ea men s o p oduc- ion animals changed a e he onse o his s udy. Dai y a me s now ha e limi ed access o an imic obials o ea some p ede ined diseases like subclinical mas i is, on ce - ain condi ions. These he ds ha e o belong o he d heal h eco ding sys em and ha e egula isi s o a supe ising e e ina ian. Mic obiological esul s o milk samples shall be a ailable be o e ea men o mas i is. The da a we e o ganized and analyzed using SAS 9.4 (SAS Ins i u e Inc., Ca y, NC, USA) and SPSS 22 (IBM SPSS S a is ics, A monk, NY, USA) so wa e. Di e ences in he ea men eco ds o mas i is be ween he s udy g oup and he con ol g oup we e es ed using a Chi squa e es . Resul s Cha ac e is ics o cows andhe ds In ou da a, 61.9% o cows we e No dic Red (NR), also known as Finnish Ay shi e, 36.6% Hols ein (HOL), and 1.8% Finnca le, Je sey, o di e en c ossb eds. The mean pa i y was 2.4 in NR and 2.2 in HOL cows ( ange 1–15). The p opo ions o pa i ies 1, 2, 3 and≥4 we e 33.8, 26.5, 18.4, and 21.3%, espec i ely. The a e age annual milk p oduc ion o he cows was 8598kg pe cow, 8323kg o NR and 9154kg o HOL cows. Mos he ds we e in con en ional p oduc ion sys ems and only 1.9% o he cows we e on o ganic a ms. 1.6% o bo h NR and HOL cows we e in o ganic p oduc ion, bu o Finnca le he p opo ion was 7.7%. The a e age he d size was 44.7, 46.5, and 48.0 in 2010, 2011, and 2012, espec i ely. The minimum numbe o cows in a he d was h ee o all yea s, bu he maximum numbe inc eased om 444 in 2010 o 498 in 2012. The p opo ions o ee- s alls and AMS inc eased du ing he s udy pe iod. In he en i e pe iod, 51.8% o he cows we e housed in ie-s all ba ns, 46.6% in di e en ypes o ee-s all ba ns, and 1.6% in ba ns wi h di e en mixed sys ems. In o al, 52.0% o he cows we e milked wi h a pipeline milking sys em in ie s alls, 30.4% in a milking pa lo , 17.0% wi h AMS, and 0.6% wi h a bucke milking machine in ie s alls. Mic obiological esul s o  he milk samples DNA o some a ge pa hogen species o g oup o spe- cies was de ec ed in 87.6% o he qua e milk samples. 12.4% o he samples we e nega i e, i.e. did no con ain DNA o any a ge species. In 49% o he samples, DNA o only one a ge species o g oup o species, and in 25.2% wo species o g oup o species, we e de ec ed. In samples wi h DNA o wo bac e ial species, a dominan Table 1 Numbe o qua e milk samples, cows sampled (s udy g oup) andcon ol cows (no milk samples submi ed) in2010, 2011 and2012 2010 2011 2012 To al numbe o samples Samples 64,630 84,413 91,024 240,067 Samples wi h one a ge pa hogen 31,731 44,037 41,743 117,511 To al numbe o cows To al numbe o indi idual cows Cows in he s udy g oup 32,723 41,949 45,582 120,254 93,529 Cows in he con ol g oup 177,409 177,205 180,427 535,041 238,235 To al numbe o cows 210,132 219,154 226,009 655,295 331,764 Page 4 o 9 Vakkamäki e al. Ac a Ve Scand (2017) 59:33 species (>90% o he o al a ge DNA in he sample) was p esen in 66.2% o he samples. In 13.4% o all samples, DNA o mo e han wo bac e ial species was ound. In hese samples, he numbe o a ge species o g oup o species de ec ed in he sample was as ollows: 3 in 9.2%, 4 in 2.9%, 5 in 0.9%, 6 in 0.3%, 7 in 0.07%, and 8 o 9 in less han 0.02%. One milk sample con ained DNA om 10 a ge species o g oup o species. The mos common bac e ia de ec ed in milk samples we e CNS and S. au eus. Resul s o milk samples wi h DNA o only one mic obial species o g oup o species a e shown in Table2. Mean cycle h eshold (C ) alues, e lec ing he amoun o pa hogen DNA in he sample, anged be ween 21.9 and 33.1 (Table2). The s aphylococ- cal BlaZ gene was ound in 27.5% o samples wi h DNA o CNS and in 26.9% o samples wi h DNA o S. au eus. In milk samples wi h wo bac e ial species o g oups o species de ec ed, wi h one dominan species, he mos common dominan species we e CNS (40.2%), S . dys- galac iae (16.7%), S. au eus (14.4%), S ep ococcus ube is (10.0%) and T. pyogenes/P. indolicus (5.3%). The mos common species p esen in he sample oge he wi h he p e iously lis ed dominan species we e C. bo is (29.6%), CNS (18.4%), yeas s (11.6%), T. pyogenes/P. indolicus (10.9%) and S. au eus (7.7%). Mas i is ea men s In o al, 53,993 cows in he s udy g oup and 63,843 cows in he con ol g oup ecei ed ea men o mas- i is, ecei ed DCT, o ecei ed bo h. The p opo ion o cows in he s udy and con ol g oups ea ed o clinical o subclinical mas i is, o wi h DCT, and he numbe o ea men s pe cow a e p esen ed in Table3. The p o- po ions o cows ea ed du ing lac a ion and wi h DCT we e signi ican ly highe in he s udy g oup han in he con ol g oup (p<0.001). Qua e milk samples o bac- e iology we e submi ed o he labo a o y on a e age om 56.4% o he cows wi h a eco ded ea men du ing lac a ion du ing he same yea ; his pe cen age inc eased du ing he s udy pe iod om 47.4 o 64.4%. P opo ions o di e en mas i is diagnoses o he s udy g oup cows among he e e ina y-supe ised mas i is ea men s epo ed in he heal h eco ding sys em in 2012 we e as ollows: acu e clinical mas i is 75.2%, sub- clinical mas i is 9.6%, ch onic mas i is 2.0%, unspeci ied mas i is 7.4%, and mas i is no ed by he d s a 5.7% (mas- i is du ing lac a ion). The espec i e igu es in he con- ol g oup we e 77.7, 8.7, 2.1, 7.4 and 4.1%. Cows in he s udy g oup and ea ed a e diagnosis o acu e clinical mas i is, had mos o en he ollowing species o g oup o species in hei milk sample du ing he ea men yea : CNS (22.6% o he cows), S. au eus (22.5%), S . ube is (9.99%), S . dysgalac iae (9.02%), E. coli (5.5%), and C. bo is (3.6%). Fo cows in he s udy g oup and ea ed o subclinical mas i is du ing he yea , he o de o he spe- cies was he same, wi h he excep ion o E. coli and C. bo is: The p opo ion o cows ha ing C. bo is was highe han ha o E. coli. The sha e o cows ha ing CNS was 31.5%. Culling o cows Al oge he 48.1% o he indi idual cows we e culled du - ing he 3-yea s udy pe iod. The eason o culling was epo ed o 88.5% o hose cows. Among he epo ed easons, he main causes we e mas i is (23.6%) and poo e ili y (19.0%). In he s udy g oup, mas i is was epo ed as he eason o culling o 49.1% o he culled cows o which he culling eason was epo ed. In he con- ol g oup, he espec i e p opo ion was 18.2%. Table4 shows he numbe s and p opo ions o cows culled due o mas i is. The pa hogens de ec ed mos o en in he milk samples o he culled cows a e also p esen ed in he able. Discussion This s udy epo s bac e iological esul s om bo ine in amamma y in ec ions using in o ma ion om mo e han 240,000 qua e milk samples analyzed wi h eal- ime PCR. Finland is o ou knowledge he only coun y Table 2 Numbe andp opo ion o samples wi hDNA o a a ge bac e ial species o g oup o species de ec ed alone in117,511 qua e milk samples analyzed in he labo a- o y o Valio L d. inFinland du ing2010–2012 using Pa h- oP oo TM Mas i is PCR Assay, and he mean cycle h esh- old (C ) alues a C alue e lec s he amoun o pa hogen DNA in he sample: he lowe C alue, he mo e DNA Pa hogen N % Mean C aluea CNS 50,836 43.3 31.5 S aphylococcus au eus 24,754 21.1 28.0 S ep ococcus ube is 10,572 9.0 26.6 S ep ococcus dysgalac iae 9292 7.9 27.0 Co ynebac e ium bo is 8510 7.2 31.9 Esche ichia coli 5562 4.7 29.3 T uepe ella pyogenes/Pep oniphilus indolicus 2921 2.5 26.0 Yeas s 1812 1.5 31.3 En e ococcus spp. 1722 1.5 30.7 Klebsiella spp. 836 0.7 28.9 S ep ococcus agalac iae 501 0.4 25.6 Se a ia ma cescens 149 0.1 33.1 P o o heca sp. 34 0 32.7 Mycoplasma spp. 7 0 21.9 Mycoplasma bo is 3 0 27.1 To al 117,511 100 Page 5 o 9 Vakkamäki e al. Ac a Ve Scand (2017) 59:33 whe e PCR es ing has la gely eplaced con en ional cul u ing in ou ine mic obiological diagnos ics o qua - e milk samples om mas i ic cows [28, 29]. The PCR me hod used he e iden i ies a wide ange o a ge pa h- ogens han commonly e ealed by ou ine con en ional cul u ing [24]. The uni o m diagnos ic me hod used and he la ge numbe s o samples allowed us o make con i- den in e ences on he mic obiological e iology o mas i- is in Finland. The majo i y o he milk samples con ained DNA o some a ge species o g oup o species. The p opo ion o nega i e a ge DNA samples, 12.4%, was lowe han he p opo ion o “no g ow h” samples in s udies using con en ional bac e iology, which gene ally has anged om 20 o 40% [13, 15, 17, 29]. This is a leas pa ly explained by he capaci y o he PCR me hod o de ec also DNA o g ow h-inhibi ed o dead mic obes [30]. The samples wi h DNA o some a ge species we e la gely easy o in e p e because in 49.0% o he samples DNA o only one a ge species o g oup o species was de ec ed. This ag ees wi h p e ious epo s whe e he same PCR es was used wi h smalle ma e ials [24, 28]. In samples wi h DNA o wo bac e ial species (25.2%), a dominan species, being he likely causal agen o IMI, was ound in 66.2% o he samples. Thus, in 75% o all samples, he causal pa hogen could be iden i ied and he esul was unequi ocal. These esul s can be conside ed diagnos i- cally alid [24]. In 13.4% o he samples, DNA o mo e Table 3 P opo ion o cows in he s udy g oup andcon ol g oup ea ed du inglac a ion o a d ying-o (d y cow he - apy) o bo h. Numbe o  ea men s pe cow du ing he s udy yea is also shown Signi icance o di e ence: a, b, c, d, e, , p<0.001 Yea Cows ea ed du inglac a ion Cows ea ed a d ying-o Cows ea ed bo hdu ing lac a ion anda d ying-o Cows, % T ea men s pe cow Cows, % Cows, % T ea men s pe cow S udy g oup 2010 31.0a1.31 9.4c6.8e2.48 2011 29.4a1.28 9.6c6.6e2.45 2012 27.7a1.27 9.2c5.8e2.40 A e age 29.3 1.29 9.4 6.4 2.44 To al numbe o ea ed cows 35,074 11,287 7632 Con ol g oup 2010 6.7b1.23 5.2d1.0 2.35 2011 5.4b1.21 5.1d0.7 2.32 2012 4.1b1.19 4.6d0.6 2.31 A e age 5.4 1.21 5.0 0.8 2.33 To al numbe o ea ed cows 28,976 30,762 4105 Table 4 Numbe and p opo ion o  he cows culled wi h mas i is as he epo ed culling eason in he s udy g oup and he con ol g oup, and he mos common pa hogens de ec ed alone in he milk samples o  he cows in he s udy g oup du ing he same yea a P opo ion o cows om he o al numbe o cows culled and culling eason epo ed in s udy g oup and in con ol g oup, espec i ely b P opo ion o cows ha ing he espec i e pa hogen in hei milk sample om he o al numbe o culled cows in s udy g oup 2010 2011 2012 Numbe o culled cows % Numbe o culled cows % Numbe o culled cows % S udy g oup 3048 50.1a4461 49.6a4599 47.9a CNS 581 19.1b949 21.3b866 18.8b S aphylococcus au eus 888 29.1b1383 31.0b1239 26.9b S ep ococcus ube is 210 6.9b336 7.5b294 6.4b S ep ococcus dysgalac iae 168 5.5b303 6.8b221 4.8b Co ynebac e ium bo is 143 4.7b225 5.0b197 4.3b Esche ichia coli 148 4.9b238 5.3b222 4.8b Con ol g oup 7825 19.9a6953 18.0a6455 16.8a Page 6 o 9 Vakkamäki e al. Ac a Ve Scand (2017) 59:33 han wo bac e ial species was de ec ed, up o en species in one sample. Cu en ly, epo ing o PCR esul s di e s om ha o con en ional cul u ing in ha all a ge spe- cies de ec ed in he sample in DNA quan i ies abo e he cycle h eshold cu -o o he es a e epo ed whe eas in cul u ing, samples wi h h ee o mo e species g ow- ing on he pla e a e disca ded, he pla e being conside ed con amina ed [26]. Spo adic colonies o mino pa hogens o common con aminan s on a pla e wi h g ow h o a majo mas i is pa hogen a e commonly igno ed [24]. The same in e p e a ion sys em could be applied also in PCR diagnos ics [31]. In gene al, PCR es ing has p o en o be a as , sensi i e and eliable es sui able o ou ine mas- i is bac e iology [24, 28, 32]. An ad an age o he PCR es is he comp ehensi e panel o pa hogens, and uni- o m p ocedu e and in e p e a ion, in con as o bac e- iological cul u ing, whe e pe o mance and p ocedu es o labo a o ies may di e [33]. High cos s o molecula me hods such as PCR may s ill limi hei use o mas i is diagnos ics in many coun ies. Ou pa hogen dis ibu ions (Table2) can be compa ed wi h hose om p e ious s udies e en hough o he s used con en ional cul u e me hods. In he p esen s udy, he mos common bac e ial species in he samples wi h DNA o only one a ge species o g oup o species we e CNS (43.3%), ollowed by S. au eus (21.1%). The p opo - ion o CNS ag ees wi h p e ious indings in he Finnish na ional su ey, whe e 49.6% o he isola es we e CNS [16], bu is much highe han in ano he Finnish s udy by Koi ula e al. [17], who epo ed ha CNS caused 18% o clinical and 24% o subclinical mas i is. In he s udy by Pi kälä e al. [16], he samples we e aken om all qua - e s o all cows in he he ds and hus he indings o bac- e ial in ec ion we e mainly associa ed wi h subclinical mas i is o ea end coloniza ion. Ou high p opo ion o CNS may pa ly be explained by he sensi i i y o he PCR assay as compa ed wi h con en ional cul u ing [24]. The high a e age C alue o samples wi h CNS, gene - ally indica ing li le DNA in he samples, also suppo s his iew (Table2). The mos common bac e ia ound in he milk sam- ples o cows wi h eco ded ea men o clinical mas i is we e CNS and S au eus, wi h equal sha es. The amoun o CNS de ec ions in milk samples (in o al) was double compa ed o ha o S. au eus de ec ions (Table2), show- ing ha mos S. au eus de ec ions bu only pa o CNS de ec ions lead o ea men o clinical mas i is. CNS a e conside ed o be mino pa hogens bu a e able o cause clinical mas i is, mos ly mild [34]. I is possible ha wi h changes in he dai y indus y CNS cu en ly cause mo e clinical mas i is han be o e, bu ano he mo e likely explana ion may be ha he d s a a e sensi i e o eac o mas i is wi h mild o ansien clinical signs, aiming o keep bulk milk SCC low. Clinical mas i is is a b oad e m ha includes all cases wi h any isible clinical sign, including mild changes in he milk appea ance o con- sis ency [35]. The high p opo ion o CNS in ou s udy was no su p ising because hey ha e become e y com- mon IMI- and mas i is-causing agen s in Finland and many o he coun ies [15, 36, 37]. In ou s udy, he p opo ion o S. au eus (21%) was highe han in he Finnish na ional su ey (10%) [16]. I was simila o he p opo ion epo ed o clinical mas i- is in Sweden [38], bu high compa ed wi h s udies om o he coun ies [12, 15, 39, 40]. The high p opo ion o S. au eus eco ded he e means ha his pa hogen con in- ues o be ex ensi ely esponsible o mas i is p oblems in Finnish dai y he ds. The p opo ions o S . ube is (9.0%), S . dysgalac iae (7.9%) and E. coli (4.7%) we e simila o he esul s o Koi ula e al. [17]. I is di icul o compa e ou p opo ion o C. bo is (7.2%) wi h ha o o he s ud- ies, in pa icula wi h hose on clinical mas i is, because C. bo is is seldom epo ed. In he na ional su ey, i s p opo ion was o e 34%, bu hen all cows and qua e s in he s udy he ds we e sampled [16]. In samples wi h wo bac e ial species wi h one dominan species he mos common concomi an species de ec ed we e C. bo is, CNS, yeas s, and T. pyogenes/P. indolicus. All hese spe- cies a e po en ial con aminan s ha o igina e om he ea end o ba n en i onmen , and hei p esence in milk samples is mos ly o no clinical impo ance. A clea di e ence be ween ou esul s and hose om o he s udies is he much highe p opo ion o en i on- men al pa hogens in many o he coun ies. In s udies om he UK, S . ube is we e he species mos commonly isola ed, ollowed by En e obac e iacae [12, 14]. In ano he coun ies, en i onmen al s ep ococci we e he mos common inding in clinical mas i is [13, 39]. S . agalac iae was a ely de ec ed in his s udy (382 cows). These samples o igina ed mainly om a ew la ge he ds: 57% o he samples we e om he ds wi h mo e han 99 cows and 27% om he ds o 69–99 cows. Many samples came om he same cows: 22% o S . agalac iae-posi i e cows had a leas wo posi i e samples (one o i e posi- i e samples) (da a no shown). The he ds wi h sampled cows we e p obably included in e adica ion p og ams. We conclude ha p opo ions o mas i is-causing bac e- ial species always di e among coun ies and he ds due o di e en en i onmen al and managemen ac o s [13, 18, 41]. The Finnish ca le heal h moni o ing sys em is mainly based on eco ds o diagnoses and ea men s made by e e ina ians [42]. In con as o he si ua ion in many o he coun ies, in Finland, as in all No dic coun ies, an imic obials a e p esc ip ion-only medicines and s ic ly con olled by e e ina ians. The cu en Finnish Page 7 o 9 Vakkamäki e al. Ac a Ve Scand (2017) 59:33 legisla ion on medical ea men o animals emphasizes he impo ance o mic obiological diagnosis be o e any an imic obial ea men o dai y ca le. In mos cases when he d s a iden i ies a cow wi h clinical mas i is hey con ac hei e e ina ian who isi s he a m. In cases o subclinical and mild clinical mas i is, he d s a i s ake a milk sample and hen con ac hei e e ina ian. Such cases do no necessa ily ecei e an imic obial ea men . In ou s udy, on a e age 35.7% o he cows in he s udy g oup and 6.2% o he con ol cows had a eco ded mas- i is ea men du ing lac a ion. The signi ican di e ence be ween he sampled s udy cows and he con ol cows was expec ed. In Finland, based on he numbe s o milk samples, o al numbe o dai y cows and an imic obial consump- ion igu es, i can be es ima ed ha a milk sample o bac e iology is aken in a majo i y o mas i is cases [43]. We ound ha qua e milk samples we e submi ed o he labo a o y on a e age o nea ly 60% and, in he las s udy yea , o e 60% o he cows wi h a eco ded ea - men du ing lac a ion. I mus be poin ed ou ha he eco ding comple eness o e e ina y-supe ised cases o mas i is in Finland has no been 100%, bu acco d- ing o a No dic s udy only 56% (95% con idence in e al 48–64%) [23]. Consequen ly, ea men eco ds may be missing om ou da a, al hough disease epo ing o dai y cows p obably has become mo e comple e since 2008 when he s udy o Wol e al. [23] was ca ied ou . Disease eco dings a e now done h ough he na ional he d heal h sys em Nase a [44] wi h which con ol o ea men s has imp o ed. Con ol cows wi h a eco ded ea men we e ei he ea ed wi hou aking a milk sample o he sample was aken bu no analyzed in he labo a o y o Valio L d. In Finland, e e ina ians some imes use ipla e aga s o apid diagnosis o acu e clinical mas i is and he esul s a e gene ally no eco ded [45]. I is s ill likely ha mas- i is cases diagnosed and sampled by a me s and ea ed ia a e e ina y p esc ip ion may emain un eco ded; his is suppo ed by he low igu e o ea men s eco ded wi h he speci ic code o no es o he d s a o ea men (da a no shown). Many cows a e epea edly ea ed. The a e age numbe o ea men s pe cow was 1.3 and ea - men s may u he accumula e o some o he cows. The numbe o cows wi h eco ded ea men s o subclinical o ch onic mas i is du ing lac a ion was low, which complies wi h he Finnish guidelines o mas i- is ea men , whe e i is gene ally ad ised o pos pone ea men o subclinical mas i is du ing lac a ion un il d ying-o , and ea men o ch onic mas i is is no ecommended [46]. Ano he possibili y is ha hese ea men s, o en gi en by he d s a a e e e ina y p e- sc ip ion, a e no epo ed. DCT was adminis e ed on a e age o 9.4% o he s udy cows and o 5.0% o con ol cows, and espec i ely 6.4 and 0.8% o he cows had been ea ed bo h du ing lac a ion and wi h DCT. Based on he consump ion igu es o in amamma y p oduc s in Finland, app oxima ely 22% o cows ecei e DCT [43]. Thus a conside able pa o he ea men eco ds o DCT a e missing om ou da a. Figu es o p opo ions o cows ea ed bo h du ing lac a ion and a d ying-o we e unexpec edly low. In gene al, DCT is ecommended o cows ha ha e had mas i is du ing lac a ion [46], bu seemingly his ad ice is no ollowed on Finnish dai y a ms. Iden i ica ion o he causal agen o IMI is impo an o mas i is managemen and ea men , bu he ou ine is s ill a e in many coun ies and he ds. Fo ins ance, in he Ne he lands, only 34% o a me s submi milk sam- ples o mic obial analysis om cows wi h clinical mas- i is and 22% om hose wi h subclinical mas i is [10]. P e en ion s a egies in he he ds should be pa hogen- speci ic because p oblems caused by di e en bac e ia need di e en app oaches [11]. T ea men o mas i is should be based on mic obiological diagnosis o a oid ine icien o unnecessa y use o an imic obials [14, 47, 48]. E en simple on- a m diagnos ics, which only clas- si ies mas i is-causing agen s in o G am-nega i e and G am-posi i e, has p o en o be use ul and educed ea men cos s [49, 50]. In he s udy g oup, mas i is was he eason o nea ly hal o he culled cows; in he con ol g oup he espec- i e p opo ion was lowe , as expec ed. Culling because o mas i is gene ally ep esen s p ema u e culling, which cons i u es one- i h o he cos s o clinical mas i is in Finland [8]. The compa ison o he sha es o he pa ho- gens in Tables2 and 4 indica es ha S. au eus was mo e common among culled cows han in he whole sample da a. Thus, p esence o S. au eus means inc eased isk o culling he cow. This may e lec he o en poo ou - come o ea men o mas i is caused by S. au eus [51]. The sha es o he es i e pa hogens p esen ed in Table4 we e simila o lowe among culled cows han in he whole da ase indica ing ha only S. au eus is a eal isk ac o o culling due o mas i is. Howe e , mo e de ailed s udies on pa hogen-speci ic diagnoses and ea men da a using ou la ge da abase will be ca ied ou . Conclusions In Finland, qua e milk samples a e ou inely aken om cows wi h mas i is. Mos a e analyzed using a PCR- es which has p o en o be an applicable me hod also o la ge-scale use in bac e ial diagnos ics. In he p e- sen s udy, mic obiological diagnosis was unequi ocal in he g ea majo i y o samples in which only one spe- cies o wo species wi h one domina ing we e de ec ed. Page 8 o 9 Vakkamäki e al. Ac a Ve Scand (2017) 59:33 The epo ing o esul s wi h mo e han wo species de ec ed should be imp o ed o make he in e p e a ion easie . CNS and S. au eus we e he mos common spe- cies de ec ed in he milk samples whe eas he p opo - ion o en i onmen al pa hogens was lowe han in many o he coun ies. S. au eus was also he mos common pa hogen among he culled cows, which emphasizes he impo ance o p e en i e measu es. Mas i is ea men s a e p obably o en based on bac e ial diagnosis, bu he ac i i y o dai y a me s o submi milk samples o bac- e ial analysis could be u he inc eased. Wi h pa ho- gen-speci ic ea men s, ine icien and unnecessa y use o an imic obials can be dec eased, which also imp o es ood sa e y. Au ho s’ con ibu ions SP, AMH and ST planned he s udy. JV, AMH, and ST pe o med analysis o he da a. All au ho s pa icipa ed in d a ing he manusc ip and ead and app o ed he inal manusc ip . Au ho de ails 1 Depa men o P oduc ion Animal Medicine, Facul y o Ve e ina y Medicine, Uni e si y o Helsinki, Pa oninkuja 20, 04920 Saa en aus, Finland. 2 Na u al Resou ces Ins i u e Finland (Luke), Koe ilan ie 5, 00790 Helsinki, Finland. Acknowledgemen s This s udy was unded by he Finnish Minis y o Ag icul u e and Fo es y, he Wal e Eh s öm Founda ion, and he Finnish Founda ion o Ve e ina y Resea ch. We also wish o hank Valio L d. o inancing he s udy and o allowing us o use hei bac e iological da abase, and P oAg ia o p o iding us da a om he Finnish dai y he d eco ding sys ems. Compe ing in e es s The au ho s decla e ha hey ha e no compe ing in e es s. Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in pub‑ lished maps and ins i u ional a ilia ions. Recei ed: 28 Oc obe 2016 Accep ed: 15 May 2017 Re e ences 1. Luke. Na u al Resou ces Ins i u e Finland. Economydoc o . S uc u al de elopmen o ag icul u e. Fo ecas o s uc u al de elopmen ; 2016a. h p://www.luke. i/en/se ices. Accessed May 2016. 2. Luke. Na u al Resou ces Ins i u e Finland. S a is ics. Numbe o Li e‑ s ock; 2016b. h p://s a .luke. i/en/numbe ‑o ‑li es ock. Accessed May 2016. 3. Manninen E. 2014 a eco d yea o milking obo sales (In Finnish). Mai o ja Me. 2015;2:55. h ps://issuu.com/mai ojame/docs/mai o‑ja‑me‑2‑2015. Accessed Jan 2016. 4. Ouwel jes W, Bee da B, Windig JJ, Calus MPL, Vee kamp RF. E ec s o managemen and gene ics on udde heal h and milk composi ion in dai y cows. J Dai y Sci. 2007;90:229–38. 5. Cicconi‑Hogan KM, Gam o h M, Riche R, Ruegg PL, S iglbaue KE, Schukken YH. Associa ions o isk ac o s wi h soma ic cell coun in bulk ank milk on o ganic and con en ional dai y a ms in he Uni ed S a es. J Dai y Sci. 2013;96:3689–702. 6. Halasa T, Huijps K, Øs e ås O, Hoge een H. Economic e ec s o bo ine mas i is and mas i is managemen : a e iew. Ve Q. 2007;29:18–31. 7. Ol enacu PA, B oom DM. The impac o gene ic selec ion o inc eased milk yield on he wel a e o dai y cows. Anim Wel . 2010;19:39–49. 8. Heikkilä AM, Nousiainen JI, Pyö älä S. Cos s o clinical mas i is wi h special e e ence o p ema u e culling. J Dai y Sci. 2012;95:139–50. 9. Ca doso CS, Hö zel MJ, Wea y DM, Robbins JA, on Keyse lingk MA. Imagining he ideal dai y a m. J Dai y Sci. 2016;99:1663–71. 10. G i ioen K, Hop GE, Hols ege MM, Vel huis AG, Lam TJ, 1Heal h4Food– Du ch Mas i is Diagnos ics Conso ium. Du ch dai y a me s’ need o mic obiological mas i is diagnos ics. J Dai y Sci. 2016;99:5551–61. 11. Ruegg P. New pe spec i es in udde heal h managemen . Ve Clin N Am‑ Food Anim P ac . 2012;28:149–63. 12. Badley AJ, Leach KA, B een JE, G een LE, G een MJ. Su ey o he inci‑ dence and ae iology o mas i is on dai y a ms in England and Wales. Ve Rec. 2007;160:253–8. 13. Olde Rieke ink RGM, Ba kema HW, Kel on DF, Scholl DT. Incidence a e o clinical mas i is on Canadian dai y a ms. J Dai y Sci. 2008;91:1366–77. 14. Milne MH, Ba e DC, Fi zpa ick JL, Biggs AM. P e alence and ae iology o clinical mas i is on dai y a ms in De on. Ve Rec. 2002;151:241–2. 15. Le ison LJ, Mille ‑Cushon EK, Tucke AL, Be ge on R, Leslie KE, Ba kema HW, e al. Incidence a e o pa hogen‑speci ic clinical mas i is on con en‑ ional and o ganic Canadian dai y a ms. J Dai y Sci. 2016;99:1341–50. 16. Pi kälä A, Ha e i M, Pyö älä S, Myllys V, Honkanen‑Buzalski T. Bo ine mas i‑ is in Finland 2001‑P e alence, dis ibu ion o bac e ia, and an imic obial esis ance. J Dai y Sci. 2004;87:2433–41. 17. Koi ula M, Pi kälä A, Pyö älä S, Män ysaa i EA. Dis ibu ion o bac e ia and seasonal and egional e ec s in a new da abase o mas i is pa hogens in Finland. Ac a Ag ic Scand A. 2007;57:89–96. 18. Ba kema HW, Schukken YH, Lam TJGM, Beiboe ML, Benedic us G, B and A. Managemen p ac ices associa ed wi h he incidence a e o clinical mas i is. J Dai y Sci. 1999;82:1643–54. 19. Du ou S, F éche e A, Ba kema HW, Mussell A, Scholl DT. In i ed e iew: e ec o udde heal h managemen p ac ices on he d soma ic cell coun . J Dai y Sci. 2011;94:563–79. 20. Piepe s S, Pee e s K, Opsome G, Ba kema HW, F ankena K, De Vlieghe S. Pa hogen g oup speci ic isk ac o s a he d, hei e and qua e le els o in amamma y in ec ions in ea ly lac a ing dai y hei e s. P e Ve Med. 2011;99:91–101. 21. Koi is o T, Seppälä J. Sa a uo a ka jan a kkailua 1898‑1998. (In Finnish). Helsinki: Maaseu ukeskus en lii on julkaisuja 929. P oAg ia Keskus en Lii o; 1998. 22. P oAg ia. Resul s o he Finnish dai y he d eco ding sys em 2015. (In Finnish); 2016. h ps://www.p oag ia. i/si es/de aul / iles/a achmen / lypsyka jan_ uo osseu annan_ ulokse _2015.pd . Accessed May 2016. 23. Wol C, Espe ed M, Lind AK, Rin akoski S, Egen all A, Lindbe g A, e al. Comple eness o he disease eco ding sys ems o dai y cows in Denma k, Finland, No way and Sweden wi h special e e ence o clinical mas i is. BMC Ve Res. 2012;8:1. 24. Hii iö H, Ri a R, Au io T, Pohjan i a T, Holopainen J, Pyö älä S, e al. Pe ‑ o mance o a eal‑ ime PCR assay in ou ine bo ine mas i is diagnos ics compa ed wi h in‑dep h con en ional cul u e. J Dai y Res. 2015;82:200–8. 25. Taponen S, Liski E, Heikkilä A‑M, Pyö älä S. Fac o s associa ed wi h in amamma y in ec ion in dai y cows caused by coagulase‑nega i e s aphylococci, S aphylococcus au eus, S ep ococcus ube is, S ep ococ- cus dysgalac iae, Co ynebac e ium bo is and Esche ichia coli. J Dai y Sci. 2017;100:493–503. 26. Hogan J, González R, Ha mon R, Nicke son S, Oli e S, Pankey J, e al. Labo a‑ o y handbook on bo ine mas i is. Madison: Na ional Mas i is Council; 1999. 27. Faba. Ve e ina y ea men codes o cows (In Finnish and Swedish). h p://www. aba. i/ i/ ie opankki/ e eys a kkailu. Accessed May 2016. 28. Koskinen MT, Wellenbe g GJ, Sampimon OC, Holopainen J, Ro hkamp A, Salmiki i L, e al. Field compa ison o eal‑ ime polyme ase chain eac ion and bac e ial cul u e o iden i ica ion o bo ine mas i is bac e ia. J Dai y Sci. 2010;93:5707–15. 29. Ve beke J, Piepe s S, Sup é K, De Vlieghe S. Pa hogen‑speci ic incidence a e o clinical mas i is in Flemish dai y he ds, se e i y, and associa ion wi h he d hygiene. J Dai y Sci. 2014;97:6926–34. 30. Taponen S, Salmiki i L, Simojoki H, Koskinen MT, Pyö älä S. Real‑ ime poly‑ me ase chain eac ion‑based iden i ica ion o bac e ia in milk samples om bo ine clinical mas i is wi h no g ow h in con en ional cul u ing. J Dai y Sci. 2009;92:2610–7. 31. Hii iö H, Simojoki H, Kalmus P, Holopainen J, Pyö älä S, Taponen S. The e ec o sampling echnique on he PCR‑based bac e iological esul s o bo ine milk samples. J Dai y Sci. 2016;99:6532–41. Page 9 o 9 Vakkamäki e al. Ac a Ve Scand (2017) 59:33 • We accep p e-submission inqui ies • Ou selec o ool helps you o ind he mos ele an jou nal • We p o ide ound he clock cus ome suppo • Con enien online submission • Tho ough pee e iew • Inclusion in PubMed and all majo indexing se ices • Maximum isibili y o you esea ch Submi you manusc ip a www.biomedcen al.com/submi Submi you nex manusc ip o BioMed Cen al and we will help you a e e y s ep: 32. Keane OM, Budd KE, Flynn J, McCoy F. Inc eased de ec ion o mas i is pa hogens by eal‑ ime PCR compa ed o bac e ial cul u e. Ve Rec. 2013;173:268. 33. Pi kälä A, Gindonis V, Wallin H, Honkanen‑Buzalski T. In e labo a o y p o iciency es ing as a ool o imp o ing pe o mance in labo a o ies diagnosing bo ine mas i is. J Dai y Sci. 2005;88:553–9. 34. Simojoki H, Salomäki T, Taponen S, Ii anainen A, Pyö älä S. Inna e immune esponse in expe imen ally induced bo ine in amamma y in ec ion wi h S aphylococcus simulans and S. epide midis. Ve Res. 2011;42:1. 35. In e na ional Dai y Fede a ion. Sugges ed in e p e a ion o mas i is e mi‑ nology. Bull In Dai y Fed. 2011:448. 36. Pyö älä S, Taponen S. Coagulase‑nega i e s aphylococci‑eme ging mas i‑ is pa hogens. Ve Mic obiol. 2009;134:3–8. 37. Tenhagen BA, Hansen I, Reinecke A, Heuwiese W. P e alence o pa ho‑ gens in milk samples o dai y cows wi h clinical mas i is and in hei e s a i s pa u i ion. J Dai y Res. 2009;76:179–87. 38. Unne s ad HE, Lindbe g A, Walle KP, Ekman T, A u sson K, Nilsson‑Ös M, e al. Mic obial ae iology o acu e clinical mas i is and agen ‑speci ic isk ac o s. Ve Mic obiol. 2009;137:90–7. 39. Sampimon O, Ba kema HW, Be ends I, Sol J, Lam TJGM. P e a‑ lence o in amamma y in ec ion in Du ch dai y he ds. J Dai y Res. 2009;76:129–36. 40. Oli ei a L, Hulland C, Ruegg PL. Cha ac e iza ion o clinical mas i is occu ing in cows on 50 la ge dai y he ds in Wisconsin. J Dai y Sci. 2013;96:7538–49. 41. B een JE, G een MJ, B adley AJ. Qua e and cow isk ac o s associa ed wi h he occu ence o clinical mas i is in dai y cows in he Uni ed King‑ dom. J Dai y Sci. 2009;92:2551–61. 42. Rin akoski S. Epidemiological e alua ion o he No dic heal h egis‑ e s o dai y cows‑da a ans e , alida ion and human in luence on disease eco dings. Helsinki: Uni e si y o Helsinki; 2013. h p://u n. i/ URN:ISBN:978‑952‑10‑9488‑0. Accessed May 2016. 43. Pyö älä S, Kaa inen L, Simojoki H, Ki ilah i‑Män ylä K. Use o in amam‑ ma y an imic obials in Finland‑ owa ds lowe consump ion and a ge ed ea men o mas i is. P oceedings XXVII Wo ld Buia ics Cong ess, June 3–8 2012, Lisbon. OC 14, p. 85. 44. ETT. Animal heal h ETT; 2017. h p://www.e . i/sisal o/e ‑english. Accessed Feb 2017. 45. E kkilä S, Simojoki H, Pi kälä A. P ak isoi a eläinlääkä i halli se‑ a päi ys ys iljelyn (In Finnish). Suom Eläinlääkä il (Finnish Ve J). 2007;113:608–12. 46. E i a. Recommenda ions o he use o an imic obial agen s in he ea ‑ men o he mos signi ican in ec ious diseases in animals. Helsinki: Finn‑ ish Food Sa e y Au ho i y E i a; 2016. h ps://www.e i a. i/globalasse s/ ie oa‑e i as a/julkaisu /julkaisusa ja /elaime /mik obilaakkeiden_kay ‑ osuosi ukse _ i‑004.pd . Accessed Oc 2016. 47. Ruegg P. Risks, eali ies and esponsibili ies associa ed wi h mas i is ea ‑ men s. In NMC Regional Mee ing, Ghen , P oceedings; 2014. p. 29–35. 48. B i en AM. The ole o diagnos ic mic obiology in mas i is con ol p o‑ g ams. Ve Clin N Am‑Food Anim P ac . 2012;28:187–202. 49. Lago A, Godden SM, Bey R, Ruegg PL, Leslie K. The selec i e ea men o clinical mas i is based on on‑ a m cul u e esul s: I. E ec s on an ibio ic use, milk wi hholding ime, and sho ‑ e m clinical and bac e iological ou comes. J Dai y Sci. 2011;94:4441–56. 50. Came on M, McKenna SL, MacDonald KA, Dohoo IR, Roy JP, Kee e GP. E alua ion o selec i e d y cow ea men ollowing on‑ a m cul u e: isk o pos cal ing in amamma y in ec ion and clinical mas i is in he subsequen lac a ion. J Dai y Sci. 2014;97:270–84. 51. Ba kema HW, Schukken YH, Zadoks RN. In i ed e iew: he ole o cow, pa hogen, and ea men egimen in he he apeu ic success o bo ine S aphylococcus au eus mas i is. J Dai y Sci. 2006;89:1877–95.