Cattle Practice 19.2

Cattle Practice NOVEMBER 2011 Volume 19 Part 2

MRSA found in dairy cows and people in Europe Holmes, M.A., University of Cambridge, Department of Veterinary Medicine, Madingley Road, Cambridge, CB3 0ES, UK ABSTRACT Staphylococcus aureus is a ubiquitous organism that causes a variety of diseases in catt le and people including mastiti s. High-level resistance to β-lactams conferred by a mecA gene encoding a modifi ed penicillin binding protein (PBP2a) was fi rst observed in the early 1960’s. These methicillin resistant S. aureus (MRSA) have been responsible for both hospital-acquired infecti ons (HA-MRSA) and more recently community-acquired MRSA (CA-MRSA). The establishment of the sequence type 398 (ST398) lineage in farm animals in the early 2000’s has provided a zoonoti c reservoir of infecti on, parti cularly in conti nental Europe, described as livestockassociated MRSA (LA-MRSA). A small number of MRSA mastiti s cases in women and outbreaks in maternity or neonatal units have been reported which are generally the result of CA-MRSA. Prior to the emergence of ST398 there have been sporadic reports of MRSA, oft en from human associated lineages, being found in bovine milk samples and cases of mastiti s. Subsequently more widespread bovine MRSA udder infecti ons have been caused by ST-398 MRSA. The recent discovery of a divergent mecA MRSA in people and dairy cows in Europe has revealed another potenti al LA-MRSA strain found in a number of S. aureus lineages (CC130, ST425, and CC1943) not previously associated with human infecti ons. The most common isolates in people and catt le being CC130 with a t843 spa-type. KEYWORDS: MRSA, Staphylococcus aureus, mastiti s INTRODUCTION Staphylococcus aureus is an important pathogen causing both hospital-acquired and communityacquired infecti ons in people and is also responsible for disease in domesti c animals. This ubiquitous organism is carried asymptomati cally in 20-30% of the human populati on (Van Belkum and others 2009). In man it causes a wide range of infecti ons including skin and soft ti ssue infecti ons (SSTI), mastiti s, bone, joint and implant infecti ons, pneumonia and septi caemia. It can be the cause of comparable disease in veterinary species but is most economically signifi cant as a cause of bovine mastiti s. Following the introducti on of penicillin in the 1940’s strains of S. aureus producing penicillinase were detected. Subsequently new β-lactam anti bioti cs unaff ected by the acti on of penicillinase (such as methicillin, oxacillin and cephalosporins) were developed. Shortly aft er the introducti on of these drugs the fi rst methicillin resistant S. aureus (MRSA) was reported in the UK (Jevons and others 1961). High-level resistance to β-lactams is conferred by a mecA gene encoding a modifi ed penicillin binding protein (PBP2a or PBP’). While mecA- or PBP2a-positi ve S. aureus are referred to as MRSA, methicillin is no longer used clinically and the majority of β-lactam suscepti bility testi ng is performed using oxacillin or cefoxiti n. The mecA gene is located in a mobile geneti c element known as a staphylococcal cassett e chromosome (SCC, SCCmec when harbouring a mecA gene). Diversity in SCCmec is refl ected in the descripti on of eleven diff erent types (I–XI). SCCmec types I, II and III are largely associated with MRSA strains that are restricted to health care infecti ons and are not generally found among the healthy populati on. These strains are oft en described as hospital-acquired or healthcare-associated MRSA (HA-MRSA) and are responsible for the epidemic nosocomial MRSA seen in many countries. The emergence of MRSA infecti ons in the 1990s in people with litt le or no contact with hospitals lead to the descripti on of community-acquired MRSA (CA-MRSA). CA-MRSA is primarily associated with SCCmec types IV and V, and also with the presence of the toxin Panton-Valenti ne leukocidin (PVL). The epidemiology of MRSA changed again with the discovery that MRSA strains with multi -locus sequence type 398 carrying SCCmec types IV and V had become established within farm animal populati ons, providing a reservoir for zoonoti c infecti ons (Smith & Pearson 2011); the term livestock-associated MRSA (LA-MRSA) has been used to describe these. STRAIN TYPING OF S. AUREUS Staphylococcus aureus is a polymorphic species with a largely clonal populati on structure (Feil and others 2003) and strain typing of isolates is a useful tool for epidemiological investi gati ons. Techniques that have been used include pulsed-fi eld gel electrophoresis (PFGE) (Tenover and others 1995), Spa-typing (Koreen and others 2004) and multi - CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 93 Peer Reviewed

locus sequence typing (MLST) (Maiden and others 1998). Pulsed-fi eld gel electrophoresis is essenti ally the comparison of large genomic DNA fragments aft er digesti on with a restricti on enzyme producing a characteristi c band patt ern on an electrophoresis gel. This technique is good for resolving strains but the band patt ern may vary from laboratory to laboratory. Spa-typing uses the DNA sequencing of a polymorphic 24-bp variable-number tandem repeat (VNTR) within the 3’ coding region of the S. aureus-specifi c staphylococcal protein A (spa) gene for discriminati on of S. aureus. A database of spatypes can be found on the RIDOM spa-type server (www.ridom.de). The S. aureus MLST scheme (Enright and others 2000) uses the allelic profi le of seven housekeeping genes to establish the strain type with a database available via the Internet for submission of strains, and interrogati on (saureus. mlst.net). Typing methods for MRSA diff er in typeability and discriminatory power. For example, ST398 is a single type based on MLST, but is nontypeable by PFGE, and can be sub-divided into multi ple types by spa-typing. Where possible both the sequence type (ST) derived from MLST and the spa-type (a number with the prefi x ‘t’) will be given in this paper. One of the advantages of MLST is that closely related strains may be recognised where they only diff er in only one of the seven alleles used. In these situati ons a group of strains may be described as a clonal complex (CC). The STs of members of a CC diff er from the parental ST in no more than one of the MLST alleles (e.g. ST130 has the following allele patt ern: arcC=6, aroE=57, glpF=45, gmk=2, pta=7, tpi=58, yqiL=52. ST1245 which is a member of CC130 has the allele patt ern: arcC=6, aroE=193, glpF=45, gmk=2, pta=7, tpi=58, yqiL=52, diff ering from ST130 only in the aroE allele). HOST SPECIFICITY Studies of S. aureus in species that carry or can be infected by S. aureus have shown that there is considerable host specifi city associated with diff erent S. aureus lineages. Bovine and human strains (identifi ed by PFGE, spa-type or MLST) carried by healthy hosts or found in disease, are normally found in one host species and tend not to be found in other host species. Each S. aureus lineage also has a characteristi c geographical distributi on with some clones being broadly distributed while others have a more limited distributi on (Monecke and others 2011). The most common CC in bovine mastiti s across herds and countries are associated with ST97, ST126, ST133, ST151, ST479 and ST771 (Smith CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 94 and others 2005, Aires-De-Sousa and others 2007, Rabello and others 2007, Sung and others 2008, Ikawaty and others 2009, Smyth and others 2009, Hata and others 2010, van den Borne and others 2010, Sakwinska and others 2011). A small proporti on of bovine isolates can be grouped with human clones, e.g. CC8, CC20 and CC25 (Monecke and others 2007, Sakwinska and others 2011, van den Borne and others 2010), demonstrati ng that host-associati on is not absolute. LIVESTOCK ASSOCIATED MRSA ΈST298) A number of reports appearing in 2004/5 suggested that the rate of carriage of MRSA was higher in people living or working on pig farms than in the wider community. Reports from the Netherlands (Voss and others 2005) and France (Armand-Lefevre and others 2005) indicated that this was primarily due to the ST398 lineage. ST398 MRSA accounted for 12% of human MRSA isolati ons recorded in 2007 in the Netherlands (van Cleef and others 2011). ST398 has recently been found outside conti nental Europe with reports of carriage in pigs and people in China (Wagenaar and others 2009), Singapore (Sergio and others 2007) and North America (Smith and others 2009). Although initi ally associated with pigs, other domesti c animal species are also aff ected including veal calves (Graveland and others 2010), dairy catt le (Vanderhaeghen and others 2010), poultry (Nemati and others 2008), dogs (Nienhoff and others 2009) and horses (van den Eede and others 2009). In additi on to ST398 one report describes a cluster of ST9 MRSA (and a single allele variant) in pigs and people in China (Wagenaar and others 2009) showing the potenti al for other lineages to be LA-MRSA. It does not appear that ST398 causes signifi cant disease in veterinary species and certainly not in pigs, the most common host species in which it is found. However, farm animals clearly represent a potenti al reservoir of infecti on for transmission to people and a challenge to disease control. MRSA AND BOVINE MASTITIS Mastiti s in dairy cows is a disease of parti cular economic importance (Seegers and others 2003). Annual rates of mastiti s may reach 100 cases/100 cow-years in some dairy herds with average rates of 30-50 cases/100 cow-years reported in many countries. In Europe, North and South America and New Zealand, S. aureus is among the commonest causes of clinical and subclinical mastiti s in dairy catt le (Bradley and others 2007, Rabello and others 2007, Petrovski and others 2009, Sampimon and others 2009).

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 95 Anti bioti c resistance patt erns normally associated with MRSA were reported in S. aureus isolates obtained from bovine mastiti s cases in Belgium in the 1970’s (Devriese and others 1972, Devriese & Hommez 1975) but without confi rmati on of the presence of a mecA gene. The strain was believed to be of human origin, although no strain typing data could be generated to confi rm that suspicion at the ti me. Several studies from Korea have identifi ed low levels of MRSA (2.5-4.5% of S. aureus isolates) in which both bovine and human MRSA lineages were found in mastiti c milk samples (Lee 2003, Kwon and others 2005, Moon and others 2007, Nam and others 2011). A report from Hungary described the isolati on of an MRSA strain from both mastiti c dairy cows and a farm worker from samples taken in 2002 (JuhaszKaszanyitzky and others 2007). The single human isolate and the 27 bovine isolates found were ST1 and spa-type t127 with a type IV SCCmec and PVL negati ve suggesti ng that transmission between host species was likely to have occurred. Although similar CA-MRSA strains have been reported in human infecti ons most of these CA-MRSA are PVL positi ve. A further ST1/t127 MRSA infecti on in a bull was reported from Switzerland (Huber and others 2010) and it is also reported to have been isolated from horses and stable workers in Austria (Cuny and others 2008). These observati ons suggest that this strain may possess characteristi cs that enable it to cross the host species barrier more readily than other lineages. A Japanese study found four ST5 MRSA isolates with SCCmec type II showing identi cal or similar geno- and serotypes to human ST5 MRSA isolates in Japan (Hata and others 2010). A similar situati on was described in Turkey where 16 MRSA were found in bovine milk samples among a sample of 93 S. aureus isolates (Turkyilmaz and others 2010). These MRSA were ST8 and CC239 (ST239 and a single allelic variant) with SCCmec types III and IV which are established human MRSA strains. Interesti ngly MRSA ST8 has also been reported in a sample from a veterinarian in Switzerland (Huber and others 2010) and in horses in Canada (Weese & van Duijkeren 2010). The patt ern of MRSA in bovine mastiti s before the emergence of LA-MRSA appears to be mainly the result of occasional and relati vely rare breakthrough of human-adapted MRSA. None of the human strains appear to have become well established in the dairy cow populati on and neither have any established bovine strains managed to acquire the mecA gene. In 2010 there were two reports of ST398 in dairy herds in Belgium and Germany. The Belgian report described 11 (9.3%) MRSA from 118 S. aureus isolates from 118 farms (Vanderhaeghen and others 2010). The strains were ST398, spatypes t011 or t567 and had SCCmec-type IVa or V, establishing that they were LA-MRSA strains of CC398 in a country where this strain was present in the pig, horse and poultry populati ons. The German study describes an investi gati on of 25 mastiti c bovine milk MRSA isolates and two human isolates, from 17 dairy farms. Three diff erent spa-types were found with t011 and a type V SCCmec in 23 isolates consistent with membership of the ST398 lineage (Fessler and others 2010). One of the other ST398 isolates had a t034 spa-type which has also been reported in S. aureus from cows in Germany and Switzerland (Monecke and others 2007). The other MRSA isolates from the German study were a CC8 strain. ST398/t034 bovine mastiti s S. aureus isolates have also been reported from Switzerland (Huber and others 2010, Huber and others 2011, Sakwinska and others 2011). Transmission of ST398 from other species to dairy cows appears to be occurring with some frequency. Although the major focus in ST398 research has been on MRSA, bovine mastiti s due to methicillin sensiti ve S. aureus (MSSA) ST398 has also been reported (van den Borne and others 2010, Sakwinska and others 2011). NOVEL MRSA EMERGING IN DAIRY COWS AND PEOPLE IN EUROPE The recent discovery of a highly divergent mecA gene possessed in a type XI SCCmec has opened up a new facet of the MRSA story. Garcia-Alvarez and others (2011) reported the investi gati on of a phenotypically methicillin resistant S. aureus isolate obtained from a bovine milk sample taken in 2007. The apparent absence of a mecA gene, when tested by PCR, was followed by whole genome sequencing in an att empt to fi nd an explanati on for the β-lactam resistance. Sequencing revealed a mecA homologue (mecALGA251, sharing a 70% identi ty with the conventi onal MRSA mecA gene) within a new SCCmec (type XI). Testi ng of 24 phenotypically similar bovine mastiti s S. aureus isolates (from a collecti on of 940 bovine mastiti s S. aureus isolates submitt ed to the Animal Health Veterinary Laboratory Agency, UK) revealed a further 13 MRSA that carried mecALGA251, arising from two diff erent lineages CC130 and ST425 (the discovery paper also included a ST151 isolate later found to have been an erroneous fi nding). A survey of human S. aureus isolates from nati onal collecti ons in the UK and Denmark identifi ed 51 further MRSA with mecALGA251, resulti ng from human carriage and from human disease. Among

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 96 the human isolates the CC130 and ST425 lineages were also found together with an additi onal CC1943 lineage not seen in catt le. Signifi cantly isolates with identi cal sequence types and spatypes were found in human and bovine isolates indicati ng potenti al transmission between the two host populati ons. ST425 isolates were found from both bovine and human samples in the southwest of England, ST1245 isolates were found in people and catt le from the east, and ST1526 isolates were found in people and catt le in the northeast (illustrated in Figure 1). Unfortunately only limited clinical informati on was available to the authors and no potenti al links between the human subjects and agriculture could be investi gated. Only human isolate collecti ons were searched in Denmark where there appears to be very litt le diversity in the new MRSA. All the Danish isolates were ST130 Figure 1. Map of England and Wales showing the approximate geographical origins of the methicillinresistant S. aureus isolates with the divergent mecA gene as reported by Garcia-Alvarez and others (2011). The cow silhouett es indicate isolates obtained from bovine samples and silhouett es of people indicate human isolates. The multi -locus sequence types (ST) are indicated by the bold text in the boxes att ached to each silhouett e and the spa types are indicated by italic text. The text boxes are grouped together to indicate the three main lineages (CC130, ST425 and CC1943).

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 97 and all but one of them were spa-type t843 (one isolate was t1535). The Danish collecti on also provided the oldest isolate with one having been collected from a bacteraemia in 1975, and as a result of almost complete collecti ons for the years 2008-10 provided evidence of a trend for increasing incidence over these years. Evidence from this study does not indicate that the prevalence of divergent mecA MRSA is high and it certainly accounts for less than 1% of human MRSA isolates in the UK and Denmark. A report from Ireland describes CC130 divergent mecA MRSA isolated from two human hospital pati ents. Both isolates were CC130. One isolate, with ST130 and spa-type t843, shares these characteristi cs with 32 isolates found in England, Scotland and Denmark (bovine and human) from the discovery study (Garcia-Alvarez and others 2011). The second isolate with ST1764 and spatype t373 shares the same sequence type but has a diff erent spa-type to the three ST1764 human isolates from Scotland. A German report describes the discovery of eleven divergent mecA MRSA isolates (Cuny and others 2011). All of the isolates were CC130 and the spa-types were t843, t1736, and t1773. The ST130/t843 appears to be the most common divergent mecA MRSA lineage throughout Europe based on the limited informati on produced to date. The ST130/t1736 strain found in Germany was also found in a human isolate from England. Personal communicati ons from Møller Aarestrup, Grundmann and Laurent indicate that the new MRSA has been found in dairy cows in Denmark and France and also in people in France and the Netherlands. The sequence type of the original bovine discovery strain was fi rst described in an atypical high-virulence S. aureus isolated from a Belgian rabbit in 1995 (Vancraeynest and others 2006). The spa-type of this isolate (PI 41/95) is t742 and so the sequence type and spa-type match a human MRSA isolate reported in the discovery paper (Garcia-Alvarez and others 2011). Should it turn out that this Belgian rabbit strain also harbours the new mecA homologue it would be interesti ng to look at the original Belgian bovine methicillin resistant S. aureus isolates described by Devriese in the 1970’s (Devriese and others 1972, Devriese & Hommez 1975) for possession of the divergent mecA gene. A study of the populati on structure of S. aureus from bovine mastiti s in the Rhone valley reported 60 Swiss, and 25 French ST151/t529 isolates (Sakwinska and others 2011) identi cal to a bovine strain that was found carrying the new mecA gene in the southwest of England (GarciaAlvarez and others 2011). However, none of the Swiss and French isolates were MRSA, and even penicillin resistance was very rare among ST151 isolates (1.2%) (Sakwinska and others 2011). CONCLUSIONS There is litt le evidence in the literature to suggest that intramammary MRSA infecti ons present clinically any diff erently to infecti ons caused by MSSA. There have been no reports of problems in dairy herds caused by MRSA where conventi onal control and treatment of mastiti s have failed. Clearly where the presence of MRSA has been confi rmed in a herd it would be wise to select an anti bioti c regime that might be expected to be eff ecti ve against methicillin-resistant organisms. Ideally this should be based on culture and sensiti vity of the strains involved. In the UK a number of products, licensed for use in bovine mastiti s, contain anti microbials that are likely to be eff ecti ve against MRSA such as pirlimycin or novobiocin (Nam and others 2011). A wider public health concern surrounds the potenti al reservoir of infecti on that dairy herds with MRSA may pose to people. Although evidence of transmission of ST398 MRSA from farm animals to man has been published there is insuffi cient evidence, at the ti me of writi ng, to be certain that transmission of the new mecA homologue MRSA from dairy cows to people is occurring. However, it would be irresponsible to assume that it does not occur unti l there is evidence to support this belief. Exposure to infected animals could lead to skin and soft ti ssue infecti ons should MRSA contaminate open wounds or abrasions. A more likely outcome is the asymptomati c carriage of MRSA by people working or living on farms which may then be passed on to the wider populati on. Further research is being performed to provide evidence concerning the risks associated with MRSA in dairy cows and to enable informed decisions to be made about public and animal health consequences of this MRSA. REFERENCES Aires-de-Sousa, M., Parente, C.E., Vieira-da-Mott a, O., Bonna, I.C., Silva, D.A., de Lencastre, H. (2007) Characterizati on of Staphylococcus aureus isolates from buff alo, bovine, ovine, and caprine milk samples collected in Rio de Janeiro State, Brazil. Appl. Environ. Microbiol. 73(12): 3845-3849 Armand-Lefevre, L., Ruimy, R., Andremont, A. (2005) Clonal comparison of Staphylococcus aureus isolates from healthy pig farmers, human controls, and pigs. Emerg. Infect. Dis. 11(5): 711-714 Bradley, A.J., Leach, K.A., Breen, J.E., Green, L.E., Green, M.J. (2007) Survey of the incidence and aeti ology of mastiti s on dairy farms in England and Wales. Vet. Rec. 160(8): 253-257 Cuny, C., Layer, F., Strommenger, B., Witt e, W. (2011) Rare Occurrence of Methicillin-Resistant Staphylococcus aureus

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 98 CC130 with a Novel mecA Homologue in Humans in Germany. PLoS One 6(9): e24360. Cuny, C., Strommenger, B., Witt e, W., Stanek, C. (2008) Clusters of infecti ons in horses with MRSA ST1, ST254, and ST398 in a veterinary hospital. Microb. Drug. Resist. 14(4): 307-310 Devriese, L.A., Hommez, J. (1975) Epidemiology of methicillin-resistant Staphylococcus aureus in dairy herds. Res. Vet. Sci. 19(1): 23-27 Devriese, L.A., van Damme, L.R., Fameree, L. (1972) Methicillin (cloxacillin)-resistant Staphylococcus aureus strains isolated from bovine mastiti s cases. Zentralbl Veterinarmed B19, 7: 598-605 Enright, M.C., Day, N.P., Davies, C.E., Peacock, S.J., Spratt , B.G. (2000) Multi locus sequence typing for characterizati on of methicillin-resistant and methicillin-suscepti ble clones of Staphylococcus aureus. J. Clin. Microbiol. 38(3): 1008-1015 Feil, E.J., Cooper, J.E., Grundmann, H., Robinson, D.A., Enright, M.C., Berendt, T., Peacock, S.J., Smith, J.M., Murphy, M., Spratt , B.G., Moore, C.E., Day, N.P. (2003) How clonal is Staphylococcus aureus? J. Bacteriol. 185(11): 3307-3316 Fessler, A., Scott , C., Kadlec, K., Ehricht, R., Monecke, S., Schwarz, S. (2010) Characterizati on of methicillin-resistant Staphylococcus aureus ST398 from cases of bovine mastiti s. J. Anti microb. Chemother. 65(4): 619-625 Garcia-Alvarez, L., Holden, M.T., Lindsay, H., Webb, C.R., Brown, D.F., Curran, M.D., Walpole, E., Brooks, K., Pickard, D.J., Teale, C., Parkhill, J., Bentley, S.D., Edwards, G.F., Girvan, E.K., Kearns, A.M., Pichon, B., Hill, R.L., Larsen, A.R., Skov, R.L., Peacock, S.J., Maskell, D.J., Holmes, M.A. (2011) Meti cillin-resistant Staphylococcus aureus with a novel mecA homologue in human and bovine populati ons in the UK and Denmark: a descripti ve study. Lancet Infect Dis. Graveland, H., Wagenaar, J.A., Heesterbeek, H., Mevius, D., van Duijkeren, E., Heederik, D. (2010) Methicillin resistant Staphylococcus aureus ST398 in veal calf farming: human MRSA carriage related with animal anti microbial usage and farm hygiene. PLoS One 5(6): e10990. Hata, E., Katsuda, K., Kobayashi, H., Uchida, I., Tanaka, K., Eguchi, M. (2010) Geneti c Variati on among Staphylococcus aureus Strains from Bovine Milk and Their Relevance to Methicillin-Resistant Isolates from Humans. J. Clin. Microbiol. Huber, H., Giezendanner, N., Stephan, R., Zweifel, C. (2011) Genotypes, Anti bioti c Resistance Profi les and Microarray-Based Characterizati on of Methicillin-Resistant Staphylococcus aureus Strains Isolated from Livestock and Veterinarians in Switzerland. Zoonoses Public Health 58(5): 343-349 Huber, H., Koller, S., Giezendanner, N., Stephan, R., Zweifel, C. (2010) Prevalence and characteristi cs of meti cillinresistant Staphylococcus aureus in humans in contact with farm animals, in livestock, and in food of animal origin, Switzerland, 2009. Euro. Surveill. 15(16): Ikawaty, R., Brouwer, E.C., Jansen, M.D., van Duijkeren, E., Mevius, D., Verhoef, J., Fluit, A.C. (2009) Characterizati on of Dutch Staphylococcus aureus from bovine mastiti s using a Multi ple Locus Variable Number Tandem Repeat Analysis. Vet. Microbiol. 136(3-4): 277-284 Jevons, M.P., Rolinson, G.N., Knox, R. (1961) “Celbenin”- resistant Staphylococci. Br. Med. J. 1: 124-125 Juhasz-Kaszanyitzky, E., Janosi, S., Somogyi, P., Dan, A., van der Graaf-van Bloois, L., van Duijkeren, E., Wagenaar, J.A. (2007) MRSA transmission between cows and humans. Emerg. Infect. Dis. 13(4): 630-632 Koreen, L., Ramaswamy, S.V., Graviss, E.A., Naidich, S., Musser, J.M., Kreiswirth, B.N. (2004) spa typing method for discriminati ng among Staphylococcus aureus isolates: implicati ons for use of a single marker to detect geneti c micro- and macrovariati on. J. Clin. Microbiol. 42(2): 792-799 Kwon, N.H., Park, K.T., Moon, J.S., Jung, W.K., Kim, S.H., Kim, J.M., Hong, S.K., Koo, H.C., Joo, Y.S., Park, Y.H. (2005) Staphylococcal cassett e chromosome mec (SCCmec) characterizati on and molecular analysis for methicillinresistant Staphylococcus aureus and novel SCCmec subtype IVg isolated from bovine milk in Korea. J. Anti microb. Chemother. 56(4): 624-632 Lee, J.H. (2003) Methicillin (Oxacillin)-resistant Staphylococcus aureus strains isolated from major food animals and their potenti al transmission to humans. Appl. Environ. Microbiol. 69(11): 6489-6494 Maiden, M.C., Bygraves, J.A., Feil, E., Morelli, G., Russell, J.E., Urwin, R., Zhang, Q., Zhou, J., Zurth, K., Caugant, D.A., Feavers, I.M., Achtman, M., Spratt , B.G. (1998) Multi locus sequence typing: a portable approach to the identifi cati on of clones within populati ons of pathogenic microorganisms. Proc. Natl. Acad. Sci. USA 95(6): 3140-3145 Monecke, S., Coombs, G., Shore, A.C., Coleman, D.C., Akpaka, P., Borg, M., Chow, H., Ip, M., Jatzwauk, L., Jonas, D., Kadlec, K., Kearns, A., Laurent, F., O’Brien, F.G., Pearson, J., Ruppelt, A., Schwarz, S., Scicluna, E., Slickers, P., Tan, H.L., Weber, S., Ehricht, R. (2011) A fi eld guide to pandemic, epidemic and sporadic clones of methicillin-resistant Staphylococcus aureus. PLoS One 6(4): e17936. Monecke, S., Kuhnert, P., Hotzel, H., Slickers, P., Ehricht, R. (2007) Microarray based study on virulence-associated genes and resistance determinants of Staphylococcus aureus isolates from catt le. Vet. Microbiol. 125(1-2): 128-140 Moon, J.S., Lee, A.R., Kang, H.M., Lee, E.S., Kim, M.N., Paik, Y.H., Park, Y.H., Joo, Y.S., Koo, H.C. (2007) Phenotypic and geneti c anti biogram of methicillin-resistant staphylococci isolated from bovine mastiti s in Korea. J. Dairy. Sci. 90(3): 1176-1185 Nam, H.M., Lee, A.L., Jung, S.C., Kim, M.N., Jang, G.C., Wee, S.H., Lim, S.K. (2011) Anti microbial suscepti bility of Staphylococcus aureus and characterizati on of methicillinresistant Staphylococcus aureus isolated from bovine mastiti s in Korea. Foodborne Pathog. Dis. 8(2): 231-238 Nemati , M., Hermans, K., Lipinska, U., Denis, O., Deplano, A., Struelens, M., Devriese, L.A., Pasmans, F., Haesebrouck, F. (2008) Anti microbial resistance of old and recent Staphylococcus aureus isolates from poultry: fi rst detecti on of livestock-associated methicillin-resistant strain ST398. Anti microb. Agents Chemother. 52(10): 3817-3819 Nienhoff , U., Kadlec, K., Chaberny, I.F., Verspohl, J., Gerlach, G.F., Schwarz, S., Simon, D., Nolte, I. (2009) Transmission of methicillin-resistant Staphylococcus aureus strains between humans and dogs: two case reports. J. Anti microb. Chemother. 64(3): 660-662 Petrovski, K.R., Heuer, C., Parkinson, T.J., Williamson, N.B. (2009) The incidence and aeti ology of clinical bovine mastiti s on 14 farms in Northland, New Zealand. NZ Vet. J. 57(2): 109- 115 Rabello, R.F., Moreira, B.M., Lopes, R.M., Teixeira, L.M., Riley, L.W., Castro, A.C. (2007) Multi locus sequence typing of Staphylococcus aureus isolates recovered from cows with mastiti s in Brazilian dairy herds. J. Med. Microbiol. 56(11): 1505-1511 Sakwinska, O., Morisset, D., Madec, J.-Y., Waldvogel, A., Moreillon, P., Haenni, M. (2011) Link between Genotype and Anti microbial Resistance in Bovine Mastiti s-Related Staphylococcus aureus Strains, Determined by Comparing Swiss and French Isolates from the Rhone Valley. Appl. Environ. Microbiol. 77(10): 3428-3432 Sampimon, O.C., Barkema, H.W., Berends, I.M., Sol, J.,

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 99 Lam, T.J. (2009) Prevalence and herd-level risk factors for intramammary infecti on with coagulase-negati ve staphylococci in Dutch dairy herds. Vet. Microbiol. 134(1-2): 37-44 Seegers, H., Fourichon, C., Beaudeau, F. (2003) Producti on eff ects related to mastiti s and mastiti s economics in dairy catt le herds. Vet. Res. 34(5): 475-491 Sergio, D.M., Koh, T.H., Hsu, L.Y., Ogden, B.E., Goh, A.L., Chow, P.K. (2007) Investi gati on of meti cillin-resistant Staphylococcus aureus in pigs used for research. J. Med. Microbiol. 56(8): 1107-1109 Smith, E.M., Green, L.E., Medley, G.F., Bird, H.E., Fox, L.K., Schukken, Y.H., Kruze, J.V., Bradley, A.J., Zadoks, R.N., Dowson, C.G. (2005) Multi locus sequence typing of interconti nental bovine Staphylococcus aureus isolates. J. Clin. Microbiol. 43(9): 4737-4743 Smith, T.C., Male, M.J., Harper, A.L., Kroeger, J.S., Tinkler, G.P., Moritz, E.D., Capuano, A.W., Herwaldt, L.A., Diekema, D.J. (2009) Methicillin-resistant Staphylococcus aureus (MRSA) strain ST398 is present in midwestern U.S. swine and swine workers. PLoS One 4(1): e4258. Smith, T.C., Pearson, N. (2011) The emergence of Staphylococcus aureus ST398. Vector Borne Zoonoti c Dis. 11(4): 327-339 Smyth, D.S., Feil, E.J., Meaney, W.J., Harti gan, P.J., Tollersrud, T., Fitzgerald, J.R., Enright, M.C., Smyth, C.J. (2009) Molecular geneti c typing reveals further insights into the diversity of animal-associated Staphylococcus aureus. J. Med. Microbiol. 58(10): 1343-1353 Sung, J.M., Lloyd, D.H., Lindsay, J.A. (2008) Staphylococcus aureus host specifi city: comparati ve genomics of human versus animal isolates by multi -strain microarray. Microbiology 154(7): 1949-1959 Tenover, F.C., Arbeit, R.D., Goering, R.V., Mickelsen, P.A., Murray, B.E., Persing, D.H., Swaminathan, B. (1995) Interpreti ng chromosomal DNA restricti on patt erns produced by pulsed-fi eld gel electrophoresis: criteria for bacterial strain typing. J. Clin. Microbiol. 33(9): 2233-2239 Turkyilmaz, S., Tekbiyik, S., Oryasin, E., Bozdogan, B. (2010) Molecular epidemiology and anti microbial resistance mechanisms of methicillin-resistant Staphylococcus aureus isolated from bovine milk. Zoonoses Public Health 57(3): 197-203 van Belkum, A., Melles, D.C., Nouwen, J., van Leeuwen, W.B., van Wamel, W., Vos, M.C., Wertheim, H.F., Verbrugh, H.A. (2009) Co-evoluti onary aspects of human colonisati on and infecti on by Staphylococcus aureus. Infect. Genet. Evol. 9(1): 32-47 van Cleef, B.A., Monnet, D.L., Voss, A., Krziwanek, K., Allerberger, F., Struelens, M., Zemlickova, H., Skov, R.L., Vuopio-Varkila, J., Cuny, C., Friedrich, A.W., Spiliopoulou, I., Paszti , J., Hardardotti r, H., Rossney, A., Pan, A., Pantosti , A., Borg, M., Grundmann, H., Mueller-Premru, M., OlssonLiljequist, B., Widmer, A., Harbarth, S., Schweiger, A., Unal, S., Kluytmans, J.A. (2011) Livestock-associated methicillinresistant Staphylococcus aureus in humans, Europe. Emerg. Infect. Dis. 17(3): 502-505 van den Borne, B.H., Nielen, M., van Schaik, G., Melchior, M.B., Lam, T.J., Zadoks, R.N. (2010) Host adaptati on of bovine Staphylococcus aureus seems associated with bacteriological cure aft er lactati onal anti microbial treatment. J. Dairy Sci. 93(6): 2550-2558 van den Eede, A., Martens, A., Lipinska, U., Struelens, M., Deplano, A., Denis, O., Haesebrouck, F., Gasthuys, F., Hermans, K. (2009) High occurrence of methicillin-resistant Staphylococcus aureus ST398 in equine nasal samples. Vet. Microbiol. 133(1-2): 138-144 Vancraeynest, D., Haesebrouck, F., Deplano, A., Denis, O., Godard, C., Wildemauwe, C., Hermans, K. (2006) Internati onal disseminati on of a high virulence rabbit Staphylococcus aureus clone. J. Vet. Med. B. Infect. Dis. Vet. Public Health 53(9): 418-422 Vanderhaeghen, W., Cerpenti er, T., Adriaensen, C., Vicca, J., Hermans, K., Butaye, P. (2010) Methicillin-resistant Staphylococcus aureus (MRSA) ST398 associated with clinical and subclinical mastiti s in Belgian cows. Vet. Microbiol. Voss, A., Loeff en, F., Bakker, J., Klaassen, C., Wulf, M. (2005) Methicillin-resistant Staphylococcus aureus in pig farming. Emerg. Infect. Dis. 11(12): 1965-1966 Wagenaar, J.A., Yue, H., Pritchard, J., Broekhuizen-Sti ns, M., Huijsdens, X., Mevius, D.J., Bosch, T., Van Duijkeren, E. (2009) Unexpected sequence types in livestock associated methicillin-resistant Staphylococcus aureus (MRSA): MRSA ST9 and a single locus variant of ST9 in pig farming in China. Vet. Microbiol. 139(3-4): 405-409 Weese, J.S., van Duijkeren, E. (2010) Methicillin-resistant Staphylococcus aureus and Staphylococcus pseudintermedius in veterinary medicine. Vet. Microbiol. 140(3-4): 418-429

An update to a practi sing vet’s view of bovine TB control in Wales Rodgers, P. J., Allen & Partners, The Veterinary Surgery, Whitland, Carmarthenshire, SA34 0QN ABSTRACT This paper discusses, from the view point of a veterinary surgeon practi sing in Wales, the control of bovine tuberculosis (bTB), updati ng a paper presented to BCVA in 2009 (Rodgers 2009). It draws from practi cal experience bTB skin testi ng over 33 years, parti cipati on in the Welsh Assembly Government (WAG) biosecurity Intensive Treatment Area (ITA) and the WG Intensive Acti on Area (IAA). The author’s involvement in the IAA is described and the impact on catt le keepers discussed. A review of skin testi ng procedures is made, with an emphasis on test performance. Data extracted from TB Master (Lilac Technology) is used to analyse previous bTB skin test results for a group of bTB reactors. KEYWORDS: Biosecurity, bovine TB, Biosecurity ITA, Wales, TB skin test, SICCT, badger cull, Intensive Acti on Area, North Pembrokeshire, TB Master INTRODUCTION The author has been in farm animal practi ce for 33 years (mainly in Whitland on the Carmarthenshire /Pembrokeshire border in West Wales). During this ti me there has been a dramati c increase in the incidence of bovine tuberculosis (bTB) in Wales, peaking in 2008-2009 when 12,043 catt le were slaughtered, costi ng £25.34m in compensati on. Although the incidence has fallen in 2010-2011 to 9078 catt le slaughtered (costi ng £12.69m) these fi gures are sti ll higher than in 2006–2007 (Welsh Government (WG) 2011). As part of the ‘One Wales’ Programme for Government, the Welsh Assembly Government (WAG) committ ed itself to “vigorously pursue a programme of bovine TB eradicati on”. On 24 March 2009 the WAG Minister for Rural Aff airs, announced her intenti on “based on comprehensive evidence, to implement a cull of badgers, alongside additi onal catt le disease control measures, within an Intensive Acti on Pilot Area (IAPA) in North Pembrokeshire to simultaneously deal with both sources of infecti on” (WAG 2009). Aft er a consultati on process the Minister was granted powers to carry out the cull. In November 2009 the Assembly voted (majority 43 to 9) to reject a proposal to stop the cull. In the same month the Badger Trust announced its intenti on to seek a judicial review of the order. The High Court ruled (April 2010) that the order was lawful (Justi ce Lloyd Jones 2010). The cull was suspended in June 2010 pending an appeal by the Badger Trust. In July 2010 the Appeal Court quashed the badger cull order on the technicality: “The respondents erred in law in making an Order for the whole of Wales having consulted on the basis of an IAPA and on the basis of evidence which at most supported culling in an IAPA.” (Lord Justi ce Pill and others 2010). In September 2010 the Minister announced her intenti on to reintroduce plans for a badger cull in north Pembrokeshire in the renamed Intensive Acti on Area (IAA). Aft er a further period of consultati on, the Minister reintroduced this plan to the WAG (March 2011), with an order more ti ghtly focussed to apply only to the IAA and not the whole of Wales. The last act of the WAG Plenary, before the 2011 electi on, was to decide that the Badger (Control Area) (Wales) Order 2011 would go ahead (majority 42 to 8). In March 2011 the animal health and welfare budgets were devolved to Wales. The Welsh Labour Party Manifesto 2011 contained a commitment to “Take a science-led approach to evaluate and review the best way of tackling Bovine TB”. Aft er the electi on of May 2011, the Welsh Labour Party made the decision to govern alone without an overall majority. Under the renamed Welsh Government (WG) the Minister for Environment and Sustainable Development has responsibility for animal health and welfare of livestock. In June 2011 the Minister announced that an independent panel of experts would be appointed by the WG’s Chief Scientifi c Adviser to review the evidence about eradicati ng bovine TB with an aim of reporti ng back in the autumn of 2011. The chairman of this panel was Professor Gaskell BVSc, PhD, DVR, MRCVS. The author will discuss the eff ects of these politi cal events on practi ce clients and the wider farming community. There will also be some personal observati ons taken from addressing a meeti ng of the pressure group Pembrokeshire Against the Cull (PAC). Since 1st May 2010 the catt le owners in the IAA have been subject to more stringent catt le control measures (compared with the rest of Wales). These will be described and the eff ect on owners discussed. There have also been biosecurity visits by practi sing veterinary surgeons to catt le owners in the IAA and a buff er zone around the IAA. The procedure for these visits will be described and CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 100

discussed. The future procurement of professional veterinary services will be put out to tender by the Animal Health Veterinary Laboratories Agency (AHVLA) in the near future. Companies successful in winning tenders for bTB testi ng work will be subject to the Health and Safety at Work Act 1974 (H&S). The operati ng procedures, as set out in the Manual of Procedures (MOP) for Local Veterinary Inspectors Secti on 1A (Defra 2002), for conducti ng a single intradermal comparati ve cervical test (SICCT), cause practi cal diffi culti es which make complying with H&S law diffi cult. The author will use a review of bTB testi ng procedures (DNV Consulti ng 2006) and 33 years of experience conducti ng bTB skin tests to discuss the practi caliti es of the test, with parti cular emphasis on test performance. TB Master (Lilac Technology) has been used by Allen & Partners in Whitland for recording of TB skin tests since 2004. An analysis of this data is used to look at previous test results for individual animals that were reactors during 2010. In conclusion the author will discuss how bTB control in Wales compares with the New Zealand “Three Legged Stool” analogy (Rodgers 2009) and if progress has been made since 2009. CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 101 INTENSIVE ACTION AREA The Intensive Acti on Area (IAA) was established by the Welsh Government (WG) to tackle all sources of bTB, in both domesti c and wild animal species by implementati on of catt le control measures, biosecurity visits and a wildlife cull. The area is primarily located in north Pembrokeshire, but includes small parts of Ceredigion and Carmarthenshire. The IAA is 288km² with a boundary of approximately 97.6km. Catt le control measures have been in place since 1st May 2010 and two biosecurity visits have been made to catt le owners in the IAA (and farms in a 2km buff er zone). At the ti me of writi ng no cull of badgers has taken place. The stati sti cs in Table 1 show that the bTB incidence in the IAA is high (39% open incidents 2010) compared with the whole of South Wales (8.5% open incidents 2010) and that this area had less net decrease in bTB incidence compared with previous years (WG 2011). Catt le Control Measures The additi onal measures to reduce the risk of disease spread by catt le in the IAA include: 1. Catt le keepers in the IAA are required to test their whole herd, including catt le in associated holdings outside the IAA every 6 months. Tests not carried out in ti me may aff ect single farm Table 1. TB incidents in the Intensive Acti on Area, 2004 to 2010. Year New Closed Net Open at year end Intensive Acti on Area 2004 41 56 -15 38 2005 53 35 18 56 2006 41 48 -7 49 2007 78 55 23 72 2008 71 63 8 80 2009 55 73 -18 62 2010 79 61 18 80 All South Wales 2004 562 545 17 392 2005 655 590 65 457 2006 634 581 53 510 2007 800 668 132 642 2008 1,005 803 202 844 2009 942 1,123 -181 663 2010 788 791 -3 660 Source: download from TB incidents database at 17 January 2011. Some results from the very late 2010 may sti ll be missing. New incidents dated by the disclosing test. Closed incidents dated by the issue of the TB10. If no TB10 has been issued the incident is assumed to be sti ll open. At January 2011 there were 309 catt le herds in the IAA and 7724 in all South Wales (WG 2011).

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 102 payments and could impact on the level of compensati on received for reactor animals (this applies to all bTB tests in Wales). 2. Removal of Sole Occupancy Authoriti es (SOAs) and implementati on of Pilot Area Sole Occupancy Authoriti es (PASOAs). The main changes introduced as part of the PASOA are that all land on a PASOA must be within the IAA and all land on a PASOA must be within 16km of the main premises. 3. Removal of all Briti sh Catt le Movement Service (BCMS) linkages between holdings inside & outside the IAA. This means that all movements in and out of the IAA must be reported to BCMS. Links between holdings inside the IAA boundary remain eligible. 4. In the event of a bovine TB herd breakdown, herds in the IAA must have two clear tests before restricti ons are lift ed. 5. New County Parish Holding (CPH) numbers or sub locati on numbers are allocated to land which crossed the IAA boundary, currently run under one CPH number. 6. All breakdowns will be traced by Animal Health and Veterinary Laboratories Agency (AHVLA) unless a veterinary risk assessment determines there is no need for tracings in individual cases. 7. Greater use of gamma interferon testi ng assessed on an individual basis. These measures have imposed on IAA catt le keepers (compared with other farmers in Wales) additi onal costs, increased administrati on ti me and more frequent handling of catt le. There have been regular meeti ngs of an IAA Industry Advisory Group (IAG) including representati ves from the WG and AHVLA, local veterinary practiti oners and prominent catt le keepers in the IAA area. It is the author’s impression, from these meeti ngs, that although initi ally the catt le measures were accepted by the catt le keepers as part of the IAA package, the conti nued inability to deliver a badger cull is straining this goodwill. Biosecurity Visits The objecti ve of the annual private veterinary surgeon visit was to play a part in the reducti on in the incidence of bTB in the IAA and to reduce the risk of introducti on of new disease. This to be achieved by the herd-keeper and private vet using their collecti ve local knowledge and experti se to work with the additi onal catt le controls and form a biosecurity acti on plan (to be reviewed annually). Although the IAA boundary is principally made up of coast, estuaries, rivers and upland open hill, some farms fall into a 2km buff er zone around the IAA. It was decided that these farms should also have a biosecurity visit. As of January 2011 there were 309 farms in the IAA and 117 in the buff er zone. The author conducted 22 biosecurity visits in the fi rst year of the IAA (between December 2009 and February 2011). Because the boundary of the IAA had not been fi nalised it was not clear at the ti me which of these farms would actually be in the area. In the second year (between December 2010 and February 2011) the author conducted 19 visits, of these 9 holdings were in the IAA and 10 were in the buff er zone (Table 2). Biosecurity risk assessment tool A biosecurity risk assessment tool (RA tool) was provided for use by the private veterinary surgeons during the biosecurity risk assessment visit. This tool was a Microsoft Excel spreadsheet, based on a similar tool used in the Biosecurity Intensive Treatment Area (ITA) in January 2007 (Van Winden 2008). There were four separate sheets in the workbook: Weighti ngs; Assessment and Acti ons; Summary; Acti ons. An expert opinion workshop was held for parti cipati ng vets to decide weighti ngs for each of the risk factors. Assessments and Acti ons Sheet Risk factors (and weighti ngs): 1. Herd size (6%) 2. Farm structure (number of blocks of land; in or out of IAA) (12%) 3. Introducti on of catt le (16%) 4. Herd mixing (grazing or housing away from home premises; att ending shows) (10%) 5. Contact with other catt le (boundaries and escapes) (6%) Table 2. Catt le numbers on farms visited by the author to conduct biosecurity risk assessments. Total Catt le Numbers Less than 10 Between 11 and 50 Between 51 and 100 Between 101 and 250 Between 251 and 500 Over 500 Total Farms Visited in Year One 5 6 0 8 21 22 Farms Visti ed in year two 3 5 1 5 32 19

6. Use of adjacent land (10%) 7. Contact with wildlife (28%) 8. Manure management (8%) 9. Shared/hired equipment (3%) 10. Cleansing and disinfecti on (1%) Each secti on was divided into sub-factors (with sub-weighti ngs). Suggested acti ons were presented under the appropriate risk factor on the same sheet of the RA tool. These acti ons carried a score (reducing the risk score) which was carried forward to the summary sheet. There was a seven questi on “Percepti ons of TB Risk” secti on to assess the catt le keepers’ confi dence and fears with regard to bTB, both generally and specifi cally to their own holding. Although these questi ons were on the Assessment and Acti ons sheet they were not part of the risk assessment and did not carry a score. Summary Sheet The summary sheet demonstrated the total risk score for each of the risk factors before and aft er the suggested acti ons (high score = high risk). Scores were given with and without herd size weighti ng and summed to give a total score for the holding. Acti ons Sheet The acti ons sheet used free text to target specifi c acti ons agreed between the catt le keeper and the vet. In the second year, acti ons from the fi rst year were recorded. If the acti ons had not been achieved fully (or not started) a reason and an esti mated ti me to completi on were given. The Visits AHVLA supplied a “pack” for each catt le keeper visit consisti ng of: • Map/maps of holding(s) • Copy of the previous year’s risk assessment (if visited in the fi rst year) • Recent test result history • Catt le Tracing System (CTS) printout • Extract from BCMS acti ons spreadsheet, giving informati on on how the land used was associated to the catt le keeper on CTS e.g. sub-locati ons, additi onal County Parish Holding (CPH) numbers • If the holding had been exempted from the additi onal IAA 6 month testi ng, an explanati on was given. Each visit was conducted in a specifi c format, starti ng with a thorough examinati on of the farm yard and buildings, with parti cular emphasis on possible access to catt le areas and feed stores by CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 103 wildlife. Farm boundaries were inspected if their eff ecti veness as a barrier to contact with catt le from other herds was not clear. The next stage was to conduct the RA using the spreadsheet tool. This included a detailed descripti on of the farm boundaries, using the maps. There were nine diff erent boundary types although the total boundary width (more or less than 3 metres) was the key factor in determining if nose to nose catt le contact was a risk. The author used a paper questi onnaire to conduct the assessment which was then copied to the Assessment and Acti ons Sheet. Although this added an extra task, it provided a robust contemporaneous record of the assessment and allowed eye contact to be maintained with the farmer during the discussion (this would have been more diffi cult if data was entered directly into the RA tool). With current risk factors recorded, there was then a discussion with the farmer around the suggested soluti ons for each of the risk factors. If these acti ons were considered (by vet and farmer) to be of suffi cient benefi t compared with the cost of implementati on they were carried forward to the acti ons sheet. Where appropriate, up to three acti ons were recorded for each year and were made specifi c to that farm. The completed RA tool and farm packs (including completed maps) were returned to AHVLA, who conducted a quality assurance (QA) audit on them. Discussion The author found the spreadsheet tool diffi cult to use compared with the ITA spreadsheet tool. The assessment and acti ons sheet was overcrowded and poorly designed. The drop-down acti ons were ambiguous for some farms. In the second year 74% of the author’s RA tools generated at least one QA query. This was in spite of considerable eff ort on the author’s part, during the farm visits, to reduce errors. This distracti on adversely aff ected the discussion between the author and farmers. The summary sheet had litt le visual impact and did not aid the process of convincing catt le keepers to make changes in farming practi ce. In the ITA Final report (Enti cott 2008) it was stated that “In order to incenti vise farmers to make changes to their biosecurity, it may be preferable to use other ways of communicati ng risks. Research shows that oft en people generally have diffi culty in interpreti ng the meaning of quanti tati ve risk assessments. Simpler ways of communicati ng risk may have more impact. These may include the use of a traffi clight system to highlight the status of diff erent risk factors. Similarly, as the ITA has shown, facilitati ng

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 104 conversati ons between farmers and vets about biosecurity risks that result in a broad list of acti on points can also have a positi ve eff ect.” The IAA RA tool made the core functi on of the Biosecurity Visit more diffi cult where it should have aided it. If more ti me and resources were allocated to soft ware design of the RA tool this would be compensated for by savings in the QA process. A full analysis of the scores for the whole IAA RAs will be undertaken by the WG; however, these are a few signifi cant points taken from the acti ons discussed with catt le keepers during the author’s 22 visits in the second year of the IAA. • Of the nine owners with holdings in the IAA four had all land within the IAA. Of the fi ve that had land both in and out of the IAA, two were no longer using the land in the IA A for grazing catt le. The additi onal catt le measures for the IAA were a factor in this change of farming practi ce. • Ten farms had “contact with other catt le” risks due to quality of fencing (barrier less than 3 metres), of these fi ve had completed work on this fencing or changed management so that nose to nose contact was unlikely. Two farms had a risk from indirect contact due to watering from streams. • Three farms had signifi cant risk from catt le introducti ons. Two of these were in a prolonged bTB breakdown and had lost their closed herd status due to replacement of reactor animals. • Twelve farms had a risk from “badger access to catt le accommodati on possible”. Two had completed work to prevent this. Four fed no concentrates. None of the keepers had seen evidence of badger access to catt le accommodati on. • “Badger access to feed stores” was possible on seven farms, two of which had completed work to prevent this risk. No owners reported signs of badger access. • All farms had a risk of “contact with wildlife at pasture”. On two farms this was the only risk factor. Two farms had stopped grazing fi elds containing badger latrines. In spite of the functi onal diffi culti es with the RA tool most of the catt le keepers engaged fully with the author’s biosecurity visit. They were genuinely interested in doing their part to reduce the incidence of bTB, but felt betrayed by the lack of a badger cull. Some believed that the cull would never take place because “politi cians are afraid of the badger groups”. There were worrying signs of stress shown by these farmers when discussing bTB, especially those who had been under a prolonged bTB breakdown. Wildlife Cull In the author’s opinion there is compelling scientifi c evidence that a badger cull in north Pembrokeshire would produce a signifi cant decrease in the incidence of catt le bTB breakdowns by reducing the opportunity for contact between infected badgers and catt le (and vice versa). Without a badger cull all sources of bTB are not being tackled in north Pembrokeshire and the principles of the IAA are not being met. A cull is unlikely in 2011 due to the delay caused by the politi cal decision to wait for the fi ndings of a review of the evidence by the expert panel. There are economic and personal impacts on any farmer under bTB breakdown. In north Pembrokeshire some farmers have an additi onal worry caused by a perceived fear of acti on by animal rights groups. This causes some farmers to be reluctant to publicly voice an opinion on the cull. Pressure groups such as Pembrokeshire Against the Cull (PAC) and the Badger Trust are vocal and well organised. In March 2010 the author was invited to address a meeti ng of PAC. In spite of being treated politely it was an inti midati ng experience. The demographic makeup of the audience was very diff erent from the farming community with which the author normally works and there was obvious potenti al for inter-community confl ict regarding this issue. It is diffi cult to believe that any logical argument could be made that would have had an infl uence on the opinions of that audience. There were many strident views expressed during the meeti ng including one by a senior member of the Badger Trust, stati ng that veterinary surgeons are not scienti sts. However, aft er the meeti ng, several members of the audience asked the author’s opinion on the risk to their own health from close contact with badgers. A REVIEW OF SKIN TB TESTING PROCEDURES In the UK the skin test used for the detecti on of bTB in live animals is the single intradermal comparati ve cervical test (SICCT). The standard operati ng procedure to conduct the SICCT is set out in the Manual of Procedures (MOP) for Local Veterinary Inspectors Secti on 1A (Defra 2002). The United Kingdom (UK) has to conform to the requirements of EU Council Directi ve 64/432/EEC (Annex B) (EU 1964). Paragraph 2.2.5.1 of this directi ve states: “Technique: Injecti on sites shall be clipped and cleansed. A fold of skin within each clipped area shall be taken between the forefi nger and thumb

and measured with callipers and recorded. The dose of tuberculin shall then be injected by a method that ensures that the tuberculin is delivered intradermically (sic). A short sterile needle, bevel edge outwards, with graduated syringe charged with tuberculin, inserted obliquely into the deeper layers of the skin may be used. A correct injecti on shall be confi rmed by palpati ng a small pea-like swelling at each site of injecti on. The skin-fold thickness of each injecti on site shall be remeasured 72 hours (± 4 hours) aft er injecti on and recorded.” In reviewing the skin bTB test procedure the author in no way condones methods that deviate from the Manual of Procedures (MOP). Readers are reminded that signing a TB52 is an act of veterinary certifi cati on. However the practi cal diffi culti es of conducti ng skin bTB tests make it diffi cult to comply with both the Manual of Procedures and the Health and Safety at Work Act 1974 (H&S) even with good handling faciliti es. With poor handling or fracti ous animals the diffi culti es are increased. In some cases the diffi culty of complying with both the MOP and H&S could delay the completi on of tests, adversely aff ecti ng bTB control. A common reason that vets have been reported for not following the MOP is when catt le fail the bTB skin test and the owner refuses to allow their removal, not because bTB was missed. Badger protecti on groups also claim that bTB skin tests are not being conducted correctly. There is a danger that the current MOP could become a charter for these groups. In 2006 Det Norske Veritas Consulti ng (DNV) were requested by the Department for Environment, Food and Rural Aff airs (Defra) and WAG to expand work they were doing on premovement bTB testi ng to include the instructi ons and interpreti ve material and their use by Local Veterinary Inspectors (LVIs). This was in response to an incident in 2005 when the owner of a Dexter bull calf identifi ed as a positi ve bTB reactor refused to allow slaughter and demanded a re-test. DNV have experti se in managing risks in areas of safety and the environment with a rule-setti ng functi on and/or determined conformance and compliance to Rules, Standards and Regulati ons. DNV produced a report “Review of TB Testi ng Procedures” (DNV Consulti ng 2006). The following are extracts from this report: “Recommendati ons: • There is an urgent need to carry out a thorough review of the present “Manual of Procedures”. Considerati on should be given to the degree to which situati onal deviati ons and mistakes can be managed out, and the CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 105 degree to which opti mising deviati ons are acceptable. It is important not only that that the test is eff ecti ve, but that LVIs conform with the procedure (once revised) to ensure that the results are defensible. It is recognised that the UK has to conform to the requirements of EU Council Directi ve 64/432/EEC which is very prescripti ve on the test protocol and may limit the opti ons for change. • SVS procedures for approval and monitoring of LVIs should be reviewed, and considerati on given to a process of ongoing monitoring or audit. • The eff ecti veness of the present IT soluti ons should be reviewed, and their use encouraged. • Considerati on should be given to ways of helping manufacturers improve the design of equipment used in TB tests, including skin measurement, combined syringe and marker and catt le crushes.” The DNV report also identifi ed two common causes for the MOP not being followed: “Many of the deviati ons result from either opti mising or situati onal errors. Opti mising errors are due to the pressures of trying to get the job done quickly; this has been exacerbated by the steady increase in herd sizes. There is oft en a need to get through a lot of animals within one day. Situati on errors arise from the conditi ons at the farm, which may include the adequacy of the handling faciliti es, the weather, and the nature of the animals. It was repeatedly pointed out that many of the conti nental cross breed animals that are now common can be very diffi cult to handle and pose a serious risk of injury to the LVI conducti ng the TB test. Faced with fracti ous animals moving about violently in the crush the LVI will try and minimise his exposure by reducing the number of ti mes he handles the animal.” Saving veterinary ti me is not the only cause of opti mising errors. Prolonged handling increases the stress to catt le which can be parti cularly serious for high yielding dairy catt le. Farmers in prolonged bTB breakdown are already dangerously stressed. The testi ng procedure, which is repeated regularly, needs to be conducted effi ciently and sympatheti cally without compromising test performance. There is a risk of injury to the veterinarian (and situati on errors) when both hands are in contact with the animal’s neck and/or equipment (scissors, callipers or TB Syringes) in close proximity to the catt le crush yoke. With some animals, this is true even when the handling faciliti es are adequate. The

risk of injury is reduced by minimising the contact ti me and/or by having only one hand at a ti me in the danger area. Animal movement and fear of injury can lead to inaccurate measurements even when the MOP is followed exactly. Test sensiti vity (Se) is the proporti on of truly diseased animals detected as positi ve in the diagnosti c assay. Test specifi city (Sp) is the proporti on of truly non-diseased animals that are correctly identifi ed as negati ve by a diagnosti c test. Test performance is defi ned here as achieving the best possible Se and Sp for the test. For the skin test this depends on the ability to detect disease by correctly identi fying positi ve reacti ons and correctly measuring their relati ve size. The best skin measurement precision that can be achieved in ideal conditi ons is plus or minus 1mm (changing a pass into a reactor under severe interpretati on). The design of the standard issue callipers reduces the scale of the measurement to half and is only marked every 2mm. It is probably more realisti c to assume plus or minus 2mm. Although bett er equipment may improve accuracy there needs to be recogniti on of the practi cal limitati ons of skin measurement. The MOP under Sub-Secti on 7(b) Day 2 states “re-measure the fold of skin at each site and record the measurement in the notebook, along with a descripti on of the type of reacti on observed”. This part of the MOP has been interpreted variously as: a requirement to measure all injecti on sites with callipers on day two even if there is no reacti on; or measure all reacti ons with callipers. The second method requires careful visual inspecti on and palpati on to detect reacti ons and, in the author’s experience, leads to a far bett er test performance than relying on skin measurement to detect reacti ons. Measuring skin where there is no reacti on is at best pointless, increases the risk of injury to personnel and produces some measurements that have to be ignored (for instance if due to skin measurement errors the second day reading is less than the fi rst day reading). It could be argued that the extra ti me and focus spent on measuring all catt le actually adversely aff ects overall test performance. Listed below are some of the additi onal steps, not required in the MOP, used by the author to increase test performance: • The use of brass “Irish design” callipers to eliminate the scale reducti on in standard callipers. • On the fi rst day of the test, the act of raising a skin fold with one hand to measure with a calliper in the other hand allows palpati on of CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 106 the skin and therefore a feel for the thickness of the skin. It is common when the calliper reading is taken that it does not agree with this impression of skin thickness. This may be due to applying abnormal pressure to the callipers or movement when taking the reading. A second reading oft en confi rms that the calliper reading was not correct. • On the second day, if the test result for an animal is a borderline one, measurement of the skin near the reacti on site is taken to check the fi rst day’s readings. If there is any doubt the more severe reading is used. In the author’s opinion multi ple skin measurements in these circumstances maximise test performance. The necessity to make professional judgements of this type should be supported in the MOP. • The use of a TB Master printout (or preferably a handheld recording device) rather than a notebook to reduce the risk of transcripti on errors and catt le identifi cati on errors. Errors or this nature are serious because they can lead to the wrong animals being removed or retained. The MOP is long overdue for re-writi ng and modernisati on in a more practi cal form with input from vets who are acti vely engaged in bTB testi ng. The present MOP does not comply with Directi ve 64/432/EEC (Annex B) (EU 1964) in every detail, for instance, “Injecti on sites shall be clipped and cleansed” (EU 1964) is not refl ected in the MOP. The prime objecti ve of the MOP should be to aid disease detecti on, not to prescripti vely comply (or over comply) with the Directi ve. Paragraph 2.2.5.2 Interpretati on of Reacti ons states “The interpretati on of reacti ons shall be based on clinical observati ons and the recorded increase(s) in skinfold thickness at the sites of injecti on 72 hours aft er injecti on of tuberculin(s)”. There should be a far greater emphasis on veterinary clinical observati ons and judgement (including epidemiology). More widespread use should be made of electronic data recording and catt le identifi cati on. The ability to analyse previous test results down to individual animals would further aid disease detecti on and decision making. ANALYSIS OF TB MASTER DATA Limitati ons Only tests conducted by Allen & Partners and recorded using TB Master are included in the data set, tests conducted by AHVLA are not included. Not all informati on recorded in TB Master is

included in the export. No att empt was made to show stati sti cal signifi cance for the results, rather to explore the potenti al for practi cal uses for bTB data handling. Method TB Master has the facility to export test data as a comma-separated values (CSV) text fi le. This data was imported into a Microsoft Access (2007) database. Entries with “null” or “0” skin measurement were detected and removed. CPH numbers were checked for consistency, in most cases to follow the format “55/***/****/01”. Duplicate entries were removed. This left 315,061 lines of data with the date range from 2004 - 2010. Expressions in queries were used to perform the following calculati ons: • Second day skin thickness minus fi rst day thickness to give bovine and avian skin thickness change. • Bovine increase minus avian increase (B-A). • To diff erenti ate results under severe interpretati on the following logical equati on was applied (using the IIf functi on): If result = “Reactor” and B-A <5 Then “Severe Reactor” and If result = “IR” and B-A <1 Then “Severe CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 107 IR”. Reacti on descripti on was not taken into account. • The age of the animals on the date of the test in days was calculated using the DateDiff functi on. This was then used to calculate an age group code. Tables for farm details, individual ear tag details, vet details and age codes were then related to the main data table. The data was then analysed using queries calculated tables and reports. Results A summary of the bTB tests conducted by Allen & Partners shows a decrease in the number of reactors and inconclusive reactors (IRs) in 2010 compared with earlier years (Table 3). In 2010 there were 444 reactors identifi ed in from 97 farm tests, involving 48 diff erent farms. Short interval tests (VE-SI) accounted for 49.7% of all tests conducted in 2010 and 82.5% of reactors. Six month tests (VE-6M) accounted for 10.3% of tests and revealed a further 7.2% of reactors (Table 3). The age group “2.5 to 5 years” had a higher percentage of reactors compared with animals tested. The age groups “3 to 12 months” and “Less Table 3. Summary of bTB Tests Conducted by Allen & Partners 2006–2010. Year No. of Farm Tests Farms Tested No. of Catt le Tests Reactors Total (% of No. Catt le Tested) I/Rs Total (% of No. Catt le Tested) (More Than 1 Test) Standard Severe Standard Severe 2010 378 174 61282 444 (0.72%) 831 (1.36%) (95) 350 (0.57%) 94 (0.15%) 590 (0.96%) 241 (0.39%) 2009 358 171 56594 606 (1.07%) 1115 (1.97%) (90) 395 (0.70%) 211 (0.37%) 656 1.16% 459 0.81% 2008 324 165 50634 636 (1.26%) 1361 (2.69%) (77) 440 (0.87%) 196 (0.39%) 850 1.68% 511 1.01% 2007 356 174 50860 582 (1.14%) 1222 (2.40%) (91) 420 (0.83%) 162 (0.32%) 708 (1.39%) 514 (1.01%) 2006 277 170 39163 448 (1.14%) 648 (1.65%) (63) 292 (0.75%) 156 (0.40%) 416 (1.06%) 232 (0.59%) A summary of bTB tests recorded using TB Master between 2006 and 2010, showing the number of farm tests conducted and the number of individual animal tests (some animals are tested more than once in the year). The number of farms that were tested and how many had more than one test in the year are also shown. The percentage fi gures for reactors and Inconclusive reactors (IRs) are against the total number of catt le tests for the year.

than 3 months” had a lower percentage of reactors compared with animals tested (Tables 4 and 5). Of the 590 tests with a standard IR result, twenty were 2x IRs and would have been removed. A further eleven catt le were IR on at least one other occasion (but not consecuti vely) and were not CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 108 Table 4. Reason for bTB Tests 2010. removed during 2010. The 94 catt le identifi ed by the data analysis as severe reactors must have had “Reactor” in the TB Master Result column and a B-A calculati on of less than fi ve (see calculati ons above). Of these severe reactors, nine were Standard IR on the test before a severe reactor result and would have been removed as 2x IRs (even if the test had been at standard interpretati on). Of the 444 reactor catt le in 2010, forty had been a standard IR once before. Of these twenty were IRs on the previous test and six were IR two tests before. The remaining fourteen were IRs between six months and six years before removal as reactors. Two catt le were IRs twice before and one animal was an IR three ti mes previously. A whole life set Test Codes Count of farm tests All Tests Reactors VE-12M 12 (3.2%) 0 VE-6M 39 (10.3%) 7 (7.2%) VE-CON 33 (8.7%) 4 (4.1%) VE-CON12 18 (4.8%) 2 (2.1%) VE-CT 4 (1.1%) 0 VE-CTNH1 1 (0.3%) 0 VE-CTNH3 4 (1.1%) 0 VE-PRMT 34 (9.0%) 0 VE-SI 188 (49.7%) 80 (82.5%) VE-TR 1 (0.3%) 0 VE-WHT 44 (11.6%) 4 (4.1%) Total 378 97 Table 5. Age Groups of animals Tested 2010. Age Groups Count of Animal Tests All Results Reactors Standard Severe All Reactors Over 5 Years 13856 (22.6%) 86 (24.6%) 33 (35.1%) 119 (26.8%) 2.5 to 5 Years 14658 (23.9%) 132 (37.7%) 41 (43.6%) 173 (39.0%) 1 to 2.5 Years 18399 (30.0%) 102 (29.1%) 16 (17.0%) 118 (26.6%) 3 to 12 Months 10562 (17.2%) 26 (7.4%) 4 (4.3%) 30 (6.8%) Less than 2 Months 3807 (6.2%) 4 (1.1%) 0 4 (0.9%) Totals 61282 350 94 444 of test results recorded using TB Master for one of these animals is shown in Table 6. UK****** 00945 was a standard IR twice in 2005. The test on10/01/06 was read under standard interpretati on, but showed twelve animals within 1mm of an IR result. The test on 07/10/08 was under severe interpretati on, revealing six standard reactors, one severe reactor, eleven standard IRs and ten severe IRs. UK****** 00945 was within 1mm of failing as a severe reactor. On 4th January 2010 it was a reactor at standard interpretati on. At post mortem there were no visible lesions (NVL). Discussion The reducti on in bTB incidence is in line with the fi gures for the rest of Wales. It should be noted that there have been cyclical changes in the incidence of bTB in the past (WG 2011). The number of VE-SI and VE-6M conducted by the practi ce refl ects the number of farms under ongoing bTB breakdown. As in previous years (Rodgers 2009) the reactors found in younger age groups represent a poor return for testi ng eff ort. Looking at whole life tests can reveal catt le that present a risk of infecti on, for instance if they have a history of multi ple IR results. The Sp for a standard interpretati on SICCT is 99.9% (de la Rua-Domenech and others 2006). The fact that UK****** 00945 failed this test on 4th January 2010 means that there is a high probability that it was infected with bTB, even though it had no visible lesions (NVL). On 7th October 2008 it passed by skin measurement (with precision of + or – 1mm), but should this animal have been taken because of its previous history? CONCLUSION In 2009 the author used an analogy used in New Zealand to explain the principles of bTB control as a “three legged stool” (Rodgers 2009). The three The percentages for the reactors are against the totals of each type of reactor.

legs are: • Testi ng and removal of infected catt le (and deer) • Control of movement from infected herds and areas of high risk of disease • Control of vector populati ons The standard interpretati on SICCT is a good herd test, controlling disease eff ecti vely in areas that do not have an infected badger populati on, but the MOP and equipment used for testi ng urgently need modernising to improve test performance and safety. More use needs to be made of data as a diagnosti c tool at herd and individual animal level. Farmers are subject to more stringent testi ng and movement restricti on (parti cularly in the IAA). There is widespread use of severe interpretati on (which has a low Sp). The use of high Se, low Sp tests in the presence of a signifi cant wildlife reservoir of infecti on will conti nue to be costly to both government and the farming industry for limited reducti on in bTB incidence. Although the fi rst two legs of the stool have been strengthened and in spite of considerable eff ort and expense on the part of the WG, the vital third leg is sti ll missing. For that reason, in the author’s opinion, bTB control in Wales is no more advanced than it was in 2009. CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 109 ACKNOWLEDGEMENTS The WG TB team, especially Richard Griffi ths, Simon Rolfe and Mark Alexander have provided useful guidance and informati on. Lilac technology made the data analysis possible and Andrew Cobner has given valuable help with database design. REFERENCES De la Rua-Domenech, R., Goodchild, A.T., Vordermier, H.M., Hewison, R.G., Christiti ansen, K.H., Clift on-Hadley, R.S. (2006) Ante mortem Diagnosis of Tuberculosis in Catt le: a Review of the Tuberculin Tests, γ-interferon Assay and other Ancillary Diagnosti c Techniques. Research in Veterinary Science 81: 190-210 Defra (2002) Manual of Procedures for Local Veterinary Inspectors. Secti on 1A. Tuberculosis-Panel 1A. DNV Consulti ng (2006) Review of TB testi ng procedures: Report for Defra and the Welsh Assembly Government report no: 22115204-02 Final Report rev 3, 12th June 2006 Enti cott , G. (2008) Evaluati on of the South Wales Biosecurity Intensive Treatment Area. Draft Final Report. EU (1964) EU Council Directi ve of 26 June 1964 on animal health problems aff ecti ng intra-Community trade in bovine animals and swine (64/432/EEC) Lord Justi ce Pill, Lady Justi ce Smith, Lord Justi ce Stanley Burnton (2010) Judgement in the Court of Appeal (Civil Division) for Case No: C1/2010/1316. Neutral Citati on Number: [2010] EWCA Civ 807. Rodgers, P.J. (2009) A Practi cing Vet’s View of Bovine TB Control in Wales. Catt le Practi ce 17: 231-246 The Hon, Mr. Justi ce Lloyd Jones (2010) Judgement in the High Court of Justi ce, Queen’s Bench Division, Swansea Civil Table 6. All test Results for a Reactor Animal in 2010. UK****** 00945 Breed: HF DOB: 21/09/1999 Female Date of Test Avian Increase Avian Reacti on Bovine Increase Bovine Reacti on B-A Result Test Reason 08-Nov-04 5 C 4 C -1 Pass VE-SI 31-Jan-05 4 C 5 C 1 I/R VE-SI 25-Apr-05 0 NIL 0 NIL 0 Pass VE-SI 01-Aug-05 3 C 4 C 1 I/R VE-SI 31-Oct-05 0 NIL 0 NIL 0 Pass VE-SI 10-Jan-06 2 C 2 C 0 Pass VE-SI 04-Jul-06 0 NIL 0 NIL 0 Pass VE-6M 25-Jun-07 0 NIL 0 NIL 0 Pass VE-12M 03-Dec-07 0 NIL 0 NIL 0 Pass VE-SI 10-Mar-08 0 NIL 0 NIL 0 Pass VE-SI 03-Jun-08 0 NIL 0 NIL 0 Pass VE-SI 07-Oct-08 2 C 2 C 0 Pass VE-SI 30-Dec-08 0 NIL 0 NIL 0 Pass VE-SI 14-Apr-09 0 NIL 0 NIL 0 Pass VE-SI 13-Jul-09 0 NIL 0 NIL 0 Pass VE-SI 19-Oct-09 0 NIL 0 NIL 0 Pass VE-SI 04-Jan-10 4 C 10 C 6 Reactor VE-SI All test results, recorded on TB Master from an animal that reacted in 2010 and had previous IR results fi ve years previously.

Justi ce Centre for Case No: CO/15769/2009. Neutral Citati on Number: [2010] EWHC 768 (Admin). Welsh Assembly Government (WAG) (2009) Consultati on on the Tuberculosis Eradicati on (Wales) Order 2009 under the Animal Health Act 1981 Welsh Government (WG) (2011) Ministerial Submission SFEJ-0033-11 Annex 6 TB Problem in Wales and Annex 23 IAA stati sti cs and epidemiological informati on. Van Winden, S. (2008) The Development of a Biosecurity Scoring Tool Focussing on the Risk of Introducti on of Bovine Tuberculosis. Catt le Practi ce 16: 131-135 CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 110

Residue surveillance in catt le O’Rourke, D., Member of the Veterinary Residues Committ ee and Ortec Consultancy, Canterbury, Kent, CT2 7RL ABSTRACT In the UK, the Veterinary Medicines Directorate (VMD) oversees the authorisati on and use of veterinary medicines in all animals. In relati on to residues, the VMD operates annual surveillance programmes to monitor various foods and animal ti ssues for the presence or absence of residues to ensure the correct use of authorised veterinary medicines and that illegal substances are not being used. This helps safeguard consumer safety. Samples are collected on farm or at the abatt oir. In 2010, a total of 34,977 samples were collected and 38,220 analyses carried out. There were 126 residues in excess of statutory Maximum Residue Limits (MRLs) or other limits (called Reference Points). Of these, 63 residues were likely to have occurred from the use of veterinary medicinal products. Sixteen (around 25%) of these residues occurred in catt le. Of concern is the increase of unacceptable residues in calf kidneys (7 cases comprising 44% of the residues in catt le) which included oxytetracycline (up to 4 ti mes the MRL), sulphadiazine (just under twice the MRL) and fl orfenicol (up to 28 ti mes the MRL). The signifi cance of these results as well as those detected in catt le kidneys and milk are discussed. Veterinary surgeons can play a key role in ensuring that the risk of residues is reduced to a minimum, parti cularly in calves which provide a disproporti onate number of non-compliant samples. CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 111

An unusual aeti ology of right-sided ‘ping’ in a lactati ng Holstein heifer Greenham, T., Passant, S.R., Davies & Johnson Veterinary Surgeons, Woodplumpton, Preston, Lancashire, PR4 0TA ABSTRACT Entrapment of the caecum by the ligamentum teres hepati s was encountered in a lactati ng Holstein heifer. This presented as a right-sided ‘ping’ in associati on with non-specifi c clinical signs. Visceral entrapments by various embryological structures have been reported previously. This case is remarkable in part because of the involvement of the caecum and in part because of the presence of the umbilical porti on of the embryological remnant. This raises questi ons regarding the incidence of persistence of the distal porti on of the falciform ligament and round ligament of the liver and the eff ect this may have on other abdominal conditi ons. It also prompts questi ons about the potenti al for movement of the caecum within the abdomen and conditi ons that predispose this occurrence. KEYWORDS: Right-sided ‘ping’, caecum, falciform ligament, round ligament, entrapment INTRODUCTION Right-sided ‘ping’ is widely reported in adult dairy cows, with a variety of aeti ologies described (Kerby 2008). The causati ve conditi on most commonly encountered in the authors’ practi ce is rightdisplacement of the abomasum (RDA), with or without volvulus. First-line treatment of rightdisplaced abomasa may be medical or surgical depending on a set of clinical criteria. Surgery tends to be undertaken on diagnosis if the animal: a) has a ‘ping’ over suffi cient area to allow reasonably high confi dence in the diagnosis, b) is physiologically stable, c) is free of concurrent disease. Animals which do not sati sfy these criteria tend to be treated medically unti l they are either clinically sound or sati sfy the criteria for surgery. The only common excepti on to this is the ‘emergency laparotomy’ where the animal does not sati sfy criterion b) but appears to have parti al or complete volvulus of the abomasum requiring urgent correcti on for survival. Occasionally, laparotomy reveals a conditi on other than RDA, usually left -displaced abomasum (LDA), intesti nal tympany or caecal dilatati on. The case detailed below featured none of these conditi ons, with laparotomy revealing an unusual clinical scenario which the authors had not encountered in practi ce or in the literature. CASE REPORT A homebred Holstein of high geneti c merit presented at 179 days in milk (DIM) in her fi rst lactati on (2 years 11 months of age). There was a history of milk-drop and inappetence lasti ng 12 hours, preceded by a 48 hour period of intermitt ent signs of mild abdominal discomfort. Previously, daily yield had averaged 34.3kg, peaking at 37.6kg (56 DIM). She had received unsuccessful services at 131 and 174 DIM. Two episodes of high somati c cell count (HSCC) had been recorded (361,000 at 90 DIM and 1,674,000 at 146 DIM). There were no other notable features in her history. Clinical examinati on revealed a slight tachycardia and tachypnoea, a moderate ketosis and a rightsided ‘ping’ over the last three rib-spaces at the midpoint of the dorsoventral plane. Rectal examinati on was unremarkable; auscultati on of the rumen indicated moti lity was reduced but not completely stati c, and left fl ank percussion was unremarkable. The animal was apyrexic. No abdominal distension was evident. A tentati ve diagnosis of right-sided abomasal dilatati on/displacement was postulated and laparotomy undertaken for confi rmati on and correcti on of this conditi on. The abdomen was entered through a right paralumbar fossa incision. A gas fi lled viscus was found extending from the caudal abdomen (directly adjacent to the incision) to the cranial abdomen (approximately 30cm posterior to the liver). This porti on of gut was constricted by a ti ght band of ti ssue that ran dorsal to the viscus causing a functi onal division with approximately 75% of the volume caudal to the constricti on. The cranial 25% of the viscus had marked gaseous distension and the resultant buoyancy was causing entrapment by the ti ssue band. Manual depression of the cranial porti on of the aff ected gut facilitated its passage caudoventral to the band, causing dispersal of the gas and allowing exteriorisati on. The structure was identifi ed as the caecum and, apart from being somewhat enlarged and fl accid, was grossly normal. The caecum was opened for drainage of the accumulated gas and fl uid, closed routi nely and restored to its normal positi on within the abdomen. Examinati on of the ti ssue band revealed a long, CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 112

tubular structure of approximately 12mm diameter extending from the umbilicus to the liver. This was presumed to be a persistence of the round ligament of the liver. The structure was dissected from its inserti on at the umbilical end to prevent recurrence of visceral entrapment and a 75mm porti on resected and submitt ed for histology. The abdomen was closed routi nely and medical treatment insti gated (glucose 40%, propylene glycol, meloxicam, metoclopramide, procaine penicillin with dihydrostreptomycin sulphate). Initi al recovery was uneventf ul, with no adverse clinical signs 48 hours post-operati vely. Nine days post-operati vely some inappetence was noted and clinical examinati on revealed a left -displaced abomasum. Given the recent surgery further invasive correcti on was contra-indicated and the animal was rolled and treated medically (glucose 40%, propylene glycol). There was transient improvement but fi ve days aft er rolling the animal relapsed with recurrence of left -displaced abomasum. This was corrected and pexied using the Grymer-Sterner technique and medical treatment conti nued (glucose 40%, propylene glycol, dexamethasone, long-acti ng ceft iofur). Following the above treatments, the heifer resumed normal acti vity within the milking group. Milk yield increased from 16kg to 40kg at 239 DIM. A 305 day yield of 9,961kg (average protein 3.18%) was achieved. HISTOLOGICAL FINDINGS Histological examinati on of transverse and longitudinal secti ons showed the structure to be comprised of several slightly irregular vessels lined by att enuated endothelial cells, surrounded by fi brous ti ssue with elongated smooth muscle fi bres. Towards the peripheral half of the wall there was less fi brous ti ssue with more extensive smooth muscle. The structure was surrounded by mesothelial cells, with one edge featuring an area of loose connecti ve ti ssue with small blood vessels. The histological diagnosis confi rmed that the structure was, as suspected, a secti on of the ligamentum teres hepati s (round ligament of the liver). DISCUSSION Visceral entrapment by various structures of embryological origin has long been recognised as being an occasional conditi on in adult catt le (Koch and others 1978, Ducharme and others 1982). Intesti nal entrapment in the umbilical porti on of the round ligament of the liver is described in detail in one study, with the authors suggesti ng a high CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 113 incidence of persistence of the falciform ligament and associated round ligament (Koch and others 1978). This supports the fi ndings in this case, where the round ligament of the liver extended to the umbilicus. Discussion with other practiti oners has prompted several other reports of a tubular structure being present between the umbilicus and the cranial abdomen, someti mes as an incidental fi nding, someti mes thought to be clinically signifi cant (Haslam, personal communicati on 2011, Swindlehurst, personal communicati on 2011) although further investi gati on was not carried out in any of these cases. This contradicts modern anatomical texts which state that only the porti on of the round ligament of the liver, which extends between liver and diaphragm, remains in the adult cow (Budras & Habel 2003). Persistence of the round ligament of the liver and falciform ligament may be of clinical relevance, not just as a source of occasional visceral entrapments, but also as a feature which may have implicati ons for other abdominal conditi ons. In the case described here there was no macroscopic sign of the falciform ligament, although the histological analysis suggests that this may have been present in the animal’s recent history. The extent of the falciform ligament’s presence in the adult cow may be perti nent in terms of its partiti oning eff ect within the abdomen. Left -displaced abomasa are common within the high-yielding dairy cow populati on. It would be interesti ng to investi gate the relati onship between incidence of persistent falciform ligaments and incidence of LDA. Similarly, if LDA does occur in animals with persistent falciform ligaments, it seems plausible that this may aff ect the ease and possibly the success rates of surgical (and nonsurgical) correcti ve techniques. It would appear that there is scope for further work to establish the incidence of persistence of these embryological structures, how much variati on there is in strength of the ti ssue, and the associati on with other clinical conditi ons. Involvement of the caecum with the round ligament of the liver is another feature of this case which raises several questi ons. It is unusual for the caecum to be present in the cranioventral positi on which must have preceded entrapment. Extending this far cranially is seen with gaseous distension but in such cases the very presence of the gas causes a buoyancy which results in the viscus remaining in the dorsal abdomen. From this it is assumed that the gaseous distension cranial to the round ligament in this case occurred aft er, rather than before, entrapment. Several mechanisms for the cranial displacement can be postulated, although

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 114 the probability of these various pathogeneses is diffi cult to evaluate. Anatomical abnormality of the caecum, small bowel or large bowel was initi ally suspected, but no evidence for this was detected during inspecti on at laparotomy. Disrupti on in caecal moti lity is commonly reported (Fubini and others 1986) and this may well have been a contributi ng factor, although most cases result in gaseous dilatati on and buoyancy rather than cranioventral displacement. Tracti on from the ileum, either directly or via the ileocaecal fold is another possibility and the previously recorded occurrences of small intesti nal entrapment (Koch and others 1978) make this plausible. However physical examinati on had revealed entrapment only of the distal ti p of the caecum with no indicati on that other structures were involved. Another theory is that the structures of the hindgut had migrated cranially to fi ll a void created by abnormal reducti on in volume of the omasum and/ or abomasum, although no such fi nding was noted during surgery. It is possible that a combinati on of these factors contributed to the abnormal positi oning of the caecum and further cases would help establish epidemiological patt erns which may give some indicati on of the pathogenesis of the conditi on. It seems likely that the LDA sustained by the animal subsequent to surgery was secondary to inappetence linked with the conditi on and the procedure itself. However, it is also possible that the two conditi ons shared some common causal factors. It has long been recognised that passage of volati le fatt y acids (VFA)s and fermentable parti culate matt er into the abomasum and the caecum is associated with reduced moti lity and gaseous distension of both of these organs (Radosti ts and others 2006), which may be signifi cant in this parti cular case. CONCLUSIONS Caecal entrapment by the round ligament of the liver is an additi on to the list of diff erenti als for right-sided ‘ping’ in adult catt le. It is evident that the round ligament of the liver and the falciform ligament remain present in a proporti on of the adult populati on and it seems likely that this conditi on is underdiagnosed due to practi cal and fi nancial considerati ons limiti ng the frequency of bovine abdominal surgery. Visceral involvement with these structures should therefore be considered in any case of non-specifi c abdominal discomfort. Presence of the round ligament of the liver and the falciform ligament may also have implicati ons on the commonly encountered LDA. There is scope for further work in this area. It is also apparent that caecal displacement in to the ventral abdomen is possible. Mechanisms for this are yet to be established sati sfactorily. The authors welcome comments on this case or reports of similar cases. ACKNOWLEDGEMENTS The authors would like to register appreciati on of their clients’ interest and support during this case. The authors would also like to thank Jane Miller of Nati onWide Laboratories for her histological interpretati on and enthusiasm for further investi gati on. Input of friends and colleagues within and outside the profession is gratefully acknowledged. REFERENCES Budras, K.D., Habel, R.E. (2003) Bovine Anatomy: an illustrated text. Schlutersche, Hannover. Ducharme, N.G., Smith, D.F., Koch, D.B. (1982) Small intesti nal obstructi on caused by a persistent round ligament of the liver in a cow. Journal of the American Veterinary Medicine Associati on 180: 1234-1236 Fubini, S.L., Erb, H.N., Rebhun, W.C. (1986) Cecal dilatati on and volvulus in dairy cow: 84 cases (1977-1983). Journal of the American Veterinary Medicine Associati on 189: 96-99 Kerby, M. (2008) Diff erenti al diagnosis and management of right-sided abdominal ‘ping’ in dairy catt le. In Practi ce 30: 98-104 Koch, D.B., Robertson, J.T., Donawick, W.J. (1978) Small intesti nal obstructi on due to persistent vitelloumbilical band in a cow. Journal of the American Veterinary Medicine Associati on 173: 197-199 Radosti ts, O.M., Gay, C.C., Hinchcliff , K.W., Constable, P.D. (10th Ed, 2006) Veterinary Medicine: a textbook of the diseases of catt le, sheep, goats, pigs and horses. Saunders Ltd, UK.

The Dairyland Initi ati ve: Wisconsin’s guide to welfare friendly dairy catt le housing Cook, N.B., Clinical Associate Professor in Food Animal Producti on Medicine, School of Veterinary Medicine, University of Wisconsin-Madison, Madison WI 53706, USA ABSTRACT Major consumer concerns exist over current dairy practi ces that relate to herd housing, herd size and ‘factory farming’, lack of grazing in our producti on systems, and the ‘hyper-producti on’ of our dairy cows. If we are to believe animal acti vists and main-stream media, large scale confi nement dairy herds are places where cows are forced to milk around the clock, where individuals are pushed to breaking point with the use of hormones and feed additi ves, resulti ng in metabolic disease, poor well-being and premature death. In our Wisconsin dairy industry, while we can fi nd examples of poor management and housing in both small and large herds, we generally fi nd excellent standards of care and well-being in our dairy herds and we work with the industry to conti nue improvement in this area. Our experiences have culminated in ‘The Dairyland Initi ati ve’ – a webbased program designed to facilitate the constructi on of welfare friendly housing for dairy catt le, which we believe is essenti al to the prosperity of our dairy industry. KEYWORDS: Welfare friendly housing, dairy catt le INTRODUCTION For years, the US dairy industry has forged ahead, becoming increasingly industrialised, adopti ng scientifi c enhancements to feed and producti vity, with the primary goal of producing a plenti ful supply of cheap, safe food at least cost. In recent years we have seen the emergence of large confi nement housed faciliti es for dairy catt le with several thousand catt le housed in one site and this has been associated with increasing concerns for animal welfare. There are a number of reasons for the growing concern over the welfare of our food-producing animals. One can look at the increasing urbanrural disconnect. Fewer families have relati ves that are involved in farming, and many are completely unaware of how meat and milk are produced. The growth in disposable income also carries with it a desire and ability to purchase bett er quality food and allows the consumer to exercise greater choice – concern over the care of the animals producing that food is central to many when they make that choice. Concerns over food safety are real, and the recent meat recalls for E.coli O157 and Salmonella issues have raised awareness and the belief that our food is no longer as safe as we once thought it was. With the industrialisati on of the dairy industry, many people fi nd themselves living near a large dairy herd (In the US referred to as a Concentrated Animal Feeding Operati on or ‘CAFO’) and one has only to read the newspaper or watch TV to ‘know’ that these animal factories are mistreati ng animals and polluti ng the environment. With such a backdrop, it is not surprising for authors such as Michael Pollan, for mainstream fi lms such as Food Inc. and for animal acti vism, led by organisati ons such as People for the Ethical Treatment of Animals (PETA) and the Humans Society of the United States (HSUS), to increasingly shed doubt on modern farming practi ces and spread the word that we should all ‘shop local’ and vegetarianism is the healthier way to live if we are to stop needless animal cruelty and global warming. Within this argument, it appears that animal housing has become ‘bad’ and something ‘unnatural’. So, are they right? Have we got it wrong? Have we developed a monster in our back yards that we need to distance ourselves from as veterinarians? Specifi cally, are we putti ng our cows in large animal factories, pushing them to metabolic breaking point with no regard for animal well-being? Or is the truth rather diff erent? In this arti cle I will focus on concerns over housing systems and what we can do to improve them. Our approach in Wisconsin is to bring the producers togetherwwithwconstructionwprofessionals, consultants and agricultural lenders and promote the concept of welfare friendly housing. This has culminated in the launch of ‘The Dairyland Initi ati ve’ last year – a web-based program to facilitate the fl ow of ideas and informati on related to housing dairy catt le. GRAZING AND ORGANIC The general public’s idealised vision of the dairy industry – one that marketers have used and misused in recent ti mes in an att empt to enhance milk sales- consists of a well managed grazing dairy herd, where cows have access to pasture almost CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 115

year round with good health and ferti lity producing ~25l of milk per cow per day predominantly from forage, organically labelled without the use of ‘dangerous’ anti bioti cs, hormones and pesti cides. Proponents of grazing point to the fact that the dairy catt le are managed in their natural state, eati ng a feed that they have been designed to harvest and magically convert into a nutriti ous food for humans. Freedom to exhibit natural behaviour is of major importance to those who wish to ensure good welfare, and grazing certainly fi ts that requirement well. There is some evidence that grazing herds have a lower prevalence of lameness, and the lower producti on may help reduce the risk for metabolic disease. Recent evidence would also suggest that allowing lame cows access to pasture might help to improve locomoti on (HernandezMendo and others 2007). Recent choice experiments where cows are allowed to decide whether to access pasture or housing have yielded some surprising results. In one study, when given the choice of housing or pasture, cows selecti vely preferred housing during the day and pasture during the night (Legrand and others 2009). In another study, cows chose to remain housed twice as oft en as moving to pasture when they were allowed the same choice aft er each milking, with high producti on level and rainfall infl uencing that choice toward housing (Charlton and others 2011). These data suggest that while cows may prefer to graze pasture at least some of the ti me, and it may be ‘good’ for them, the choice may not be as important to them as we may wish to make it from our preconceived noti ons. Grazing isn’t without its problems. Environmental regulati ons would not allow all herds to graze because of the risk of manure run off contaminati ng waterways. The lower producti on from grazing, if taken to a nati onal scale, would take us back to the 1940s and the market for milk would be such that we would need many more catt le to sustain the producti on required to meet the demand. Many of the same people who feel good about grazing catt le are also the same people that are trying to reduce their carbon footprint and help prevent global warming. Finding that the US current carbon footprint for the dairy industry is 37% less than it was in the 1940s may surprise some and perhaps paint a diff erent picture (Capper and others 2009). Given that the Food and Agriculture Organisati on of the United Nati ons (FAO) warns that we will need 100% more feed to provide to the 9 million people who will inhabit the planet in 2050, it is hard to believe that this increased producti on is going to come from an industry that returns to grazing. CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 116 Finally, anyone involved in managing dairy herds at grass knows that they are not without their problems. Grass staggers in the spring, parasiti sm, ketosis in early lactati on and lameness – especially when young-stock transiti on from pasture to the milking herd at the ti me of fi rst calving - and where track maintenance is poor causing injuries as cows walk great distances to and from the milking centre, can all be serious welfare issues. Dealing with inclement weather during the winter can present a major challenge for udder health and the provision of adequate feed year round can in some situati ons be problemati c – leading to what can only be described as controlled starvati on. Heat stress in the summer can also be a major concern in some climates where there is inadequate shade. For these reasons it is common for grazing herds to construct winter housing to confi ne the animals in the winter – creati ng a facility that they use only part of the year, with many of the limitati ons that we complain about in herds that house year-round. Because organic marketi ng is so ti ghtly associated with grazing, it is someti mes hard to separate the two. However, organic is a management choice, not a housing choice - in fact, Langford and others (2008) have shown that there is no diff erence in the quality of housing between organic and non-organic herds. Some organic dairy herds have excellent standards of management and care, however some do not. Like some conventi onal herds, they can be poorly managed and badly housed. In fact there is litt le scientifi c evidence for organic herds having improved health and welfare (Huxley and others 2004), although one study recently showed a lower prevalence of lameness (Rutherford and others 2009). The organic movement’s desire to promote their product as anti bioti c and hormone free is also damaging to the industry as a whole as it implies that conventi onal milk is somehow contaminated. If we follow the science, it clearly shows that there is no signifi cant diff erence in the hormonal content of milk from organic dairies and conventi onal dairy herds that either use or do not use BST. Some organic milk actually contains signifi cantly more oestrogen and progesterone than conventi onal milk (Vicini and others 2008). In short, organic does not mean bett er or healthier; in my view, it just means the management used to produce the milk was diff erent and followed a diff erent philosophy, which consumers should be allowed to support, provided they are aware of the truth. The vast majority of grazing herds also uti lise winter housing for half of the year, so this management approach does not preclude the need to improve the quality of housing and reduce

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 117 any potenti al negati ve impact on welfare. CONFINEMENT FACTORY FARMING By defi niti on in the US, any dairy herd with more than 700 cows is a CAFO (a Concentrated Animal Feeding Operati on) typically with all-year-round total confi nement of animals in large freestall barns with a milking parlour operati ng around the clock. This would fi t many people’s defi niti on of a ‘factory farm’, even though the majority are owned by a single or group of families and hence, by defi niti on, are family farms. The term CAFO is actually used to place more stringent controls on manure and waste handling and clean water management for large livestock units. Depending on the day, if we visit such a farm we may fi nd the problems that our confi nement housed dairy industry commonly faces – too many dead cows, too many lame cows, and fresh cows suff ering from retained placenta, metriti s, and displaced abomasum. A 3% death rate looks very diff erent on a 50 cow dairy compare to a 5,000 cow dairy on any given day. These producti onrelated diseases may be associated by some with confi nement and the ‘hyperproducti on’ induced by geneti c manipulati on, producti on enhancers and geneti cally modifi ed feeds, and used as an example of why we should move away from this style of producti on system. We can certainly fi nd associati ve evidence in the scientifi c literature that cows with high milk producti on are more suscepti ble to subacute ruminal acidosis, lameness, inferti lity and metabolic disease to ‘prove’ this point. Without a doubt, there is signifi cant concern expressed by the general public regarding the size of these producti on units. In trying to determine an answer to the questi on ‘How big is too big?’ I made an assessment of the factors that impact a cow’s daily ti me budget that are infl uenced by herd size (Cook 2008). These include: 1. Milking ti me and ti me away from the pen, reducing the ti me available for rest 2. Walking distance to and from the parlour resulti ng in excessive hoof wear, thin soles and lameness 3. Our ability to detect a sick fresh cow in a pen of other fresh cows leading to unrecognised health problems that are left untreated The impact of the fi rst two factors can be lessened by milking with a larger parlour, and compressing the fl oor-plan of the barn using mechanical venti lati on systems. However, it is diffi cult to overcome the limitati ons of trying to fi nd a sick cow in a pen of fresh cows that gets larger and larger as herd size increases. Such a limitati on puts a limit on herd size of ~2,000 cows by my esti mati ons, if we are to preserve 12h per day for rest and maintain excellent standards of fresh cow health. This number is a lot larger than the answer most people who ask me the questi on have in their own mind, and it is not to say that there aren’t many well-run dairy herds larger than this. It does however make people stop and think and perhaps consider that bigger isn’t always bett er, and that there are limiti ng factors associated with herd size, but perhaps these factors come into play at herd sizes much larger than we have traditi onally been comfortable with. In 2010, from our Wisconsin Dairy Herd Improvement Associati on (DHIA) records system, there are 697 herds larger than 150 cows on test. The largest herd on test has 6,508 lactati ng cows. For the most part, we assume that these herds are predominantly freestall housed and of these herds, 99 achieve a rolling herd average milk producti on greater than 13,000 litres. They are the elite of the industry and we have had the pleasure of working with 42 of these herds in recent years. 88% of these herds uti lise sand for bedding their freestalls and they achieve excellent levels of performance (Table 1). While performance does not always equate to good welfare, the very fact that these herds exist would suggest that the percepti on that large dairy herds are by defi niti on bad for cows is erroneous. Some would look at the producti on of these herds and suggest that these cows are geneti c monsters bred to be producti on units – frail animals pushed to their limits, living on a knifeedge and on the brink of metabolic exhausti on. Table 1. Performance of elite herds with more than 150 milking cows (assumed to be freestall housed) with rolling herd average milk producti on greater than 13,000 litres. Herd Size Transiti on Cow Index Weighted Annual Mean SCC Age at 1st Calving Days Dry Turnover Rate Mean 691 848 205 24 57 36 Range 158-6508 -944-2590 80-407 21-28 36-79 20-58

That is what the animal acti vist groups would like to have us think, but we do not believe that this is the case. It is true to say that in the mid-90s our geneti c indices favoured selecti on for producti on over conformati on and health/fi tness traits, but that is no longer true. Figure 1 shows the current (2009) weighti ng of geneti c indices around the world for Holstein catt le. The US ranks 3rd and 5th in the world for indices that use the most conformati on and health/fi tness traits, bested only by Scandinavia and the Netherlands. The UK is 7th on the list. If we take the ti me to visit these elite US herds, we generally fi nd that behind these numbers are animals that are well-cared for, with excellent standards of health and welfare. Almost uniformly, these herds have lameness under control and excellent cow comfort and owners that are passionate about the health of their stock. THE WISCONSIN BLUEPRINT Our research on these high producing dairy herds would suggest that there are environmental and management changes that can be made that will reduce the risk for the common producti on related diseases. We believe that they are not an inevitable consequence of milking the kind of cow that we have bred, but a failure to provide her with the CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 118 things she needs to be successful. Improvements in stall design lead to increased milk producti on and a reducti on in lameness. Provision of enough bunk space for all of the pre-fresh and fresh cows to eat at the same ti me has reduced the incidence of ketosis, displaced abomasum and other periparturient diseases. Use of positi ve pressure venti lati on systems reduces the incidence of respiratory disease in youngstock. Our research on the risk factors associated with mastiti s, lameness, calf morbidity and fresh cow health has culminated in the creati on of a set of guidelines for housing and management – The Wisconsin Blueprint. Rather than try to come up with a set of ‘ideal’ plans that farmers can att empt to duplicate, we have tried to sell a set of ideas and principles that farmers and constructi on professionals can use to build into their own ideas and create a facility that works best for their catt le and for them. The aims of the Wisconsin Blueprint are to: 1. Accommodate cows and calves in groups which are socially stable over ti me, and manage groups to minimise movements between them 2. Wherever possible provide exposure to natural light and venti lati on, but to uti lise mechanical assistance where needed Figure 1. Relati ve contributi on of producti on, conformati on and health/fi tness traits to geneti c selecti on indices around the world in 2009.

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 119 3. Provide a comfortable place to rest for a target period of 12 hous per day (h/d) for both lame and non-lame cows which is designed to meet the space requirement of the animal, and not inhibit rising or lying movements 4. Provide enough feed and water space for each animal to opti mise metabolic health 5. Design barn layouts that do not result in undue ti me out of the pen and minimise trauma to the cows’ feet 6. Minimise morbidity and mortality rates To achieve these goals, we have incorporated as many science based recommendati ons as we can fi nd to create ‘biological’ standards to use alongside engineering standards and practi ces in the constructi on of superior dairy catt le faciliti es that are both economically viable and welfare friendly. We now have many examples of well-managed confi nement housed freestall dairy herds where producti on and welfare are harmonious – to the benefi t of the producer and the cow. In October 2010, we launched a web-site ‘htt p:// thedairylandinitiative.vetmed.wisc.edu/’wthat incorporates the Wisconsin Blueprint along with virtual tours of faciliti es of which we have assisted in the design and creati on. From the farm offi ce, producers and their consultants can tour diff erent faciliti es and use the blueprint to answer questi ons that they have related to the design of dairy catt le housing. We have also used the program to create dialogue between agricultural lenders and constructi on professionals in the State to learn from mistakes that have been made and fi ne-tune the advice being given. Because it is a web-site, the updates can be rapid and we use Facebook and Twitt er to noti fy those interested of changes that are made. In order to make it self-sustaining, the site is restricted access and there is a fee of $100 for a 2-year subscripti on. However, we have also obtained sponsorship to make access free to farmers, extension workers and veterinarians in Wisconsin. The web-site is available to subscribers world-wide. CONCLUSION My belief is that dairy cows benefi t from improvements to housing to enhance welfare in a variety of diff erent management systems, and The Dairyland Initi ati ve is being used to make these enhancements. In the future we must conti nue to questi on what we are doing and provide farmers soluti ons to promote economic survival and at the same ti me ensure that our catt le are well cared for. Consumers have diverse tastes and viewpoints and we will not please all of the people all of the ti me. Farmers should be able to graze their catt le if they choose, and people should be able to buy organic milk if they wish to do so. At the same ti me, the dairy industry has to be accountable and create improvements where it is needed. We must questi on how big a herd is ‘too big’, what kind of lying surface is best for our cows, what pen design is opti mal for transiti on cows and how best to venti late our calf barns. The Dairyland Initi ati ve is providing a forum for that discussion by soliciti ng input from farmers, consultants, builders and lenders. ACKNOWLEDGEMENTS The Dairyland Initi ati ve is funded by the Ira and Ineva Baldwin Wisconsin Idea Endowment and sponsored by commercial companies listed at http://thedairylandinitiative.vetmed.wisc.edu/ prv_supporters.htm REFERENCES Capper, J.L., Cady, R.A., Bauman, D.E. (2009) Environmental impact of dairy producti on: 1944 compared with 2007. J Anim. Sci. 87: 2160-2167 Charlton, G.L., Rutt er, S.M., East, M., Sinclair, L.A. (2011) Preference of dairy cows: Indoor cubicle housing with access to a total mixed rati on vs. access to pasture. Appl. Anim. Behav. Sci. 130: 1-9 Cook, N.B. (2008) Designing welfare friendly housing for dairy cows. Proc. Am. Assoc. Bov. Pract. Conf. 41: 78-84 Hernandez-Mendo, O., von Keyserlingk, M.A.G., Veira, D.M., Weary, D.M. (2007) Eff ect of pasture versus freestall housing on lameness in dairy cows. J. Dairy Sci. 90: 1209-1214 Huxley, J.N., Burke, J., Roderick, S., Main, D.C.J., Whay, H.R. (2004) Animal welfare assessent benchmarking as a tool for health and welfare planning in organic herds. Vet. Rec. 155: 237-239 Langford, F.M., Rutherford, K.M.D., Jack, M.C., Sherwood, L., Lawrence, A.B., Haskell, M.J. (2009) A comparison of management practi ces, farmer perceived disease incidence and winter housing on organic and non-organic dairy farms in the UK. J. Dairy Res. 76: 6-14 Legrand, A.L., von Keyserlingk, M.A.G., Weary, D.M. (2009) Preference and usage of pasture versus free-stall housing by lactati ng dairy catt le. J. Dairy Sci. 92: 3651-3658 Rutherford, K.M.D., Langford, F.M., Jack, M.J., Sherwood, L., Lawrence, A.B., Haskell, M.J. (2009) Lameness prevalence and risk factors in organic and non-organic dairy herds in the United Kingdom. The Vet. J. 180: 95-105 Vicini, J., Etherton, T., Kris-Etherton, P., Ballam, J., Denham, S., Staub, R., Goldstein, D., Cady, R., McGrath, M., Lucy, M. (2008) Survey of retail milk compositi on as aff ected by label claims regarding farm-management practi ces. J. Am. Diet. Assoc. 108: 1198-1203

Anti microbial selecti on for calf pneumonia Pott er, T.J., Veterinary Surgeon and Consultant, Westpoint Veterinary Group, Dawes Farm, Bognor Road, Warnham, West Sussex, RH12 3SH ABSTRACT The wide use of anti microbial drugs in the control of catt le diseases has led to concerns over the potenti al for emergence of resistance and the associated risks to human health. In response to these concerns and to maintain the effi cacy of the available products it is important that veterinary surgeons develop an evidence based approach to their use of anti microbials in conditi ons such as calf pneumonia, taking into account knowledge of a drug’s pharmacokineti cs and pharmacodynamics. KEYWORDS: Calf, pneumonia, anti microbial, anti bioti c, pharmacokineti cs, pharmacodynamics INTRODUCTION Anti microbial drugs are widely used in the treatment of bacterial infecti ons in animals as well as humans. Their safe and eff ecti ve use is criti cally dependent on selecti ng an appropriate drug, in an appropriately formulated product and its administrati on at an opti mal dose rate and dose interval. However, some infecti ons are becoming untreatable and reports of hospital infecti ons with methicillin resistant Staphylococcus aureus (MRSA) have heightened public concern. Isolati on of MRSA from animals was fi rst reported in 1972 following its detecti on in milk from mastiti c cows (Devriese and others 1972). Since then there have been occasional reports of MRSA infecti on in domesti c species, but in recent years the number of cases seems to be increasing (O’Mahony and others 2005). Although at present the role of domesti c animals in the epidemiology of MRSA in humans is poorly understood, the emergence of non-typeable MRSA (NT-MRSA) in the Netherlands and its link with pig and potenti ally catt le farming (van Loo and others 2007) highlights the need for responsible use of anti microbials in veterinary species. Resistance diminishes the eff ecti veness of therapy and leads to the spread of infecti ous diseases. One major reason for increasing levels of resistance is selecti on from the populati on of bacteria of those which have developed or carry genes for resistance. The high level of usage of anti microbials has been suggested as a major driver for this. In 2008 total sales of anti microbials for therapeuti c use in veterinary species in the UK was 384 tonnes, approximately 85% (327 tonnes) of which was desti ned for use in food producing animals (Goodyear 2009). Figure 1 shows the breakdown of total sales by class. The use of anti microbials in animals is both necessary and justifi ed but drugs should be uti lised CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 120 Figure 1. Proporti on of 2008 UK anti microbial sales by class. Total sales for 2008 were 384 tonnes.

in such a way as to minimise the potenti al selecti on pressure for resistance. The outcome of therapy depends on two major factors; 1. The Drug Pharmacokineti cs – achieving penetrati on of the drug to the infecti on site in a suffi cient concentrati on and for a suffi cient period of ti me. 2. The Drug Pharmacodynamics – the potency and effi cacy of the drug against the target microorganisms at the site of infecti on. When considering drug selecti on and dosing schedules, the complex interacti ons between drug pharmacokineti cs (PK) and pharmacodynamics (PD) must be considered along with factors relati ng to the animal and the disease in order to decide on appropriate therapy (Figure 2). CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 121 Pharmacokineti cs describes the absorpti on into and the fate of the drug within the host body while pharmacodynamics describes the acti on of the drug on the target, in the case of anti microbial drugs, on the pathogen populati on. Important pharmacodynamic parameters that can be determined in vitro include the Minimum Inhibitory Concentrati on (MIC) and Minimum Bactericidal Concentrati on (MBC). MIC is the concentrati on of anti microbial drug (μg/mL) that, under standard conditi ons, completely inhibits bacterial growth in the logarithmic growth phase. MBC is the concentrati on of anti microbial drug that reduces the populati on size by 99.9%. MIC and MBC vary between both species and strains of microorganisms. Growth inhibiti on curves, also called ti me-kill curves, may also be studied in vitro to give informati on on the speed of onset and durati on of the anti microbial eff ect. The MICs can be used together with pharmacokineti c data to determine eff ecti ve and Figure 2. Factors to consider in the rati onal design of dosing schedules. opti mal dosage regimens. Pharmacodynamics may initi ally be determined in vitro in broth; however, the acti on of a drug as well as bacterial growth may be aff ected by the diff erent conditi ons present in vivo. In veterinary medicine the most well-established approach for dose determinati on is the doseti trati on trial (Riviere and others 2009) which consists of testi ng diff erent dose levels and selecti ng one based on pre-established requirements in terms of cure rate. Limitati ons with the doseti trati on approach have lead to the investi gati ons of alternati ve approaches such as the use of Pharmacokinetic/Pharmacodynamicw(PK/PD) integrati on and modelling (Lees and others 2004). In the PK/PD approach the plasma concentrati on is assumed to control the ti me course of the drug at the site of acti on and thus the plasma concentrati on ti me profi le is used as the variable to explore the dose-eff ect relati onship (not the nominal administered dose). In the case of calf pneumonia the data for the pharmacokineti cs is increasingly being based on the drug concentrati on at the site of infecti on (lung ti ssue, Pulmonary Epithelial lining fl uid (PELF) etc). When opti mising the dosage regimen for anti microbial drugs there are two key outcomes: (1) Opti misati on of clinical effi cacy and (2) Minimisati on of the selecti on and spread of resistant pathogens. Three PK/PD indices have been established to predict anti microbial eff ecti veness: the area under the curve (AUC)/MIC rati o, the Cmax/MIC rati o and ti me greater than MIC; all these indices are surrogate markers for clinical effi cacy and bacterial eradicati on. Considerati ons for calf pneumonia Calf pneumonia can be caused by a number of pathogens. Viral agents include bovine respiratory syncyti al virus (BRSV), parainfl uenza III (PI3) virus, infecti ous bovine rhinotracheiti s (IBR) and bovine viral diarrhoea virus (BVDV), and it is oft en suggested that these along with the mycoplasmal agents (Mycoplasma bovis, Mycoplasma dispar and Ureaplasma spp) are the primary infecti ous agents (Andrews 2004). With the excepti on of certain instances of BRSV, fatality is usually not associated with viral infecti ons alone, instead, their roles are primarily to assist in establishing a respiratory environment that is favorable to colonisati on and replicati on by several pathogenic bacteria resulti ng in pneumonia (Panciera & Confer 2010) commonly isolated bacteria include; Mannheimia haemolyti ca, Pasteurella multocida and Histophilus somni.

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 122 Eff ecti ve treatment and control of respiratory disease is determined by rapid and accurate identifi cati on of disease. All too frequently there are delays in identi fying animals requiring treatment or an acceptance of a low level of chronic disease within a group. Studies in abatt oirs have shown high incidence of lung lesions in animals with no history of being identifi ed as having had, or being treated for respiratory disease. In one study, 37% of animals with no recorded history of respiratory disease showed lung lesions at slaughter: such levels of disease would have undoubtedly had an impact on performance. Producers need to realise that eff ecti ve treatment must begin with a commitment to accurate and early identifi cati on of sick animals. Given the large number of pathogens that have been implicated in calf pneumonia, treatment tends to be symptomati c with a broad spectrum of eff ect, with the main component of any protocol being the use of anti microbial drugs. When considering which anti microbial to use, there is a wide selecti on of products at the vet’s disposal; in fact there are more anti microbial drugs licensed for the management of respiratory disease than for any other disease of catt le. There are a number of key considerati ons when prescribing anti microbial therapy: 1. Selecti ng an anti microbial agent with an appropriate spectrum of acti vity. 2. Using a dosage protocol that att ains and maintains an eff ecti ve therapeuti c concentrati on at the site of infecti on. 3. Treati ng for an appropriate durati on. Unfortunately in the majority of cases vets will be prescribing in the absence of bacteriology and pathogen identifi cati on, so the majority of treatment decisions will be made upon previous experience on the farm and reported suscepti bility patt erns. In the absence of data specifi c to the outbreak being dealt with it is wise to select an anti microbial agent with proven effi cacy against the most commonly encountered bacterial respiratory pathogens. It should be remembered that these are biological systems and not all calves will respond to therapy in the same way. Most calves will respond well to appropriate therapy administered in a ti mely manner for an appropriate ti me period; but there will be outliers. Some animals will seemingly respond rapidly and completely to therapy because they would have recovered despite treatment. Unfortunately, at the other end of the spectrum there will be a percentage of animals that will not respond well to treatment because of other factors such as poor immune functi on. The adopti on of herd health planning and the implementati on of treatment protocols for specifi c diseases can help ensure that disease is rapidly identifi ed by the farmer and that the fi rst line therapy is appropriate. Treatment protocols for calf pneumonia, like any other disease, should be evidenced based and adapted for individual farms taking into account the principles of prudent use of anti microbials alongside the knowledge of the PK and PD of the drugs to be used. ACKNOWLEDGEMENTS The work presented was undertaken as part of the authors PhD which was funded by Defra. REFERENCES Andrews, A.H. (2004) Calf respiratory disease. Bovine Medicine. Oxford, UK. Blackwell Science Ltd. Devriese, L.A., Van Damme, L.R., Fameree, L. (1972) Methicillin (cloxacillin)-resistant Staphylococcus aureus strains isolated from bovine mastiti s cases. Zentralbl Veterinarmed B 19(7): 598-605 Goodyear, K. (2009) Sales of anti microbial products authorised for use as veterinary medicines anti protozoals anti fungals and coccidiostats in the UK in 2008, The Veterinary Medicines Directorate. Lees, P., AliAbadi, F.S., Toutain, P.L. (2004) PK-PD Modelling: An alternati ve to dose ti trati on studies for anti microbial drug dosage selecti on. Regulatory Aff airs Journal. 175-180 O’Mahony, R., Abbott , Y., Leonard, F.C., Markey, B.K., Quinn, P.J., Pollock, P.J., Fanning, S., Rossney, A.S. (2005) Methicillinresistant Staphylococcus aureus (MRSA) isolated from animals and veterinary personnel in Ireland. Veterinary Microbiology 109(3-4): 285-296 Panciera, R.J., Confer, A.W. (2010) Pathogenesis and Pathology of bovine pneumonia. Veterinary Clinics of North America Food Animal Practi ce 26(2): 191-214 Riviere, J.E., Papich, M.G. (2009) Veterinary pharmacology and therapeuti cs. Ames, Iowa, Wiley-Blackwell. van Loo, I., Huijsdens, X., Tiemersma, E., de Neeling, A., van de Sande-Bruinsma, N., Beaujean, D., Voss, A., Kluytmans, J. (2007) Emergence of methicillin-resistant Staphylococcus aureus of animal origin in humans. Emerg. Infect. Dis. 13(12):

Finding soluti ons to lameness problems in dairy herds Cook, N.B., Clinical Associate Professor in Food Animal Producti on Medicine, School of Veterinary Medicine, University of Wisconsin-Madison, Madison WI 53706, USA ABSTRACT Foot lameness is dominated by infecti ous lesions such as digital dermatiti s and foot rot (phlegmon), and by lesions of the claw horn, such as white line disease and ulcerati on of the sole. Infecti ous hoof disease can be eff ecti vely controlled through good nutriti on, improved foot hygiene and by the eff ecti ve use of hoofb aths to clean and disinfect the foot on a regular basis. Claw horn lesions may be triggered through poor feeding and poor transiti ons at calving ti me. However, white line disease is exacerbated in faciliti es with poor fl ooring that creates risk for trauma, wear and concussion. Strategic use of rubberised surfaces and improved concrete fi nishing greatly impacts the incidence of this disease. Increased standing ti me per day on hard surfaces appears to increase the risk for sole ulcer. Dairy faciliti es need to be designed so that both lame and nonlame cows can achieve at least 12 hours of rest per day. This can be achieved with deep loose-bedded stalls with sand bedding, sized to accommodate the resti ng area of the cow, with freedom to front lunge in the stall without obstructi on. Group sizes need to be matched to parlour size and throughput, in order to minimise ti me out of the pen to less than one hour per milking, and stocking rates need to be controlled to a maximum of 1.2 cows per stall. Strategic use of fans and soakers, or other cooling methods are essenti al components of lameness control in the summer. Finally, excellent hoof health requires a dedicated, well-trained hooftrimmer rebalancing the claws of all the cows at least twice per year, and prompt eff ecti ve treatment of the lame cow. KEYWORDS: Lameness, soluti ons, dairy catt le INTRODUCTION Lameness can systemati cally undermine the management of the dairy herd. No other disease has such fundamental and extensive eff ects on producti on, reproducti on and risk of early herd removal. It is the most important conditi on impacti ng a cow’s well-being, and it aff ects everything she does in her day, from resti ng, to eati ng, socialising and milking. While interest and concern for dairy cow lameness have grown in the last decade world-wide, there has been litt le improvement in the prevalence of lameness despite growing understanding of the role of important risk factors such as management, environment and nutriti on. Lameness prevalence is sti ll reported to be 15-40% for the average confi nement housed dairy herd. If anything, the situati on has become worse as herds have increased in size and cows spend more ti me on concrete surfaces. So, how do we address this problem? I will contend that we currently know enough to solve the vast majority of lameness problems in the dairy industry. What we seem to lack is a systemati c approach to identi fying the soluti ons we need and a mechanism to convey that message to the farmer. Over the last decade we have developed systems to troubleshoot herd health problems at the University of Wisconsin-Madison. My experiences troubleshooti ng lameness problems, largely in confi nement housed dairy herds are the subject of this arti cle. They have also been commercially packaged in an all-encompassing lameness troubleshooti ng and preventi on program marketed by Zinpro Corporati on as ‘First Step™’. LOCOMOTION OR MOBILITY SCORING Lameness prevalence is commonly assessed using a 4 or 5-point system scoring all cows or representati ve groups of cows in the herd. If all of the cows cannot be scored, groups should be chosen to represent fi rst lactati on heifers, fresh cows, early lactati on mature cows, late lactati on cows and the sick or lame cow pen. Scoring is usually performed as cows exit the parlour using the following rules: • The cows should be scored while walking on a solid, fl at, non-slip, well-lit surface. If repeated observati ons are made, the same surface should be used. • If possible, the herd should be scored with a secondary observer in order to remove some of the subjecti vity of the scoring. • It is oft en benefi cial to have the dairy owner or manager present for at least part of the scoring process. Such an assessment is used to report the proporti on of cows at each score, NOT the mean score for the herd, with a goal of <15% cows scoring >2 on either the 4 or 5-point system. This assessment can then be used to esti mate milk yield loss and the associated fi nancial loss of lameness. CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 123

HOOF TRIMMING ASSESSMENT Observati on of the work of the hoof-trimmer is an essenti al part of the lameness investi gati on. While there are many excellent hoof trimmers who do a great deal of good preventi ng lameness, there are unfortunately poorly trained individuals who actually do harm to the cows that they trim. If this is occurring it must be recognised and stopped as soon as possible if other improvements to lameness management on the farm are to have any eff ect. It is important that hoof trimmers visit the farm frequently enough to cope with the requested number of trims per year, and that they do not rush. An ‘expected number of trims’ based on the size of the herd, the stated frequency of trimming (for example twice a lactati on), whether or not heifers are trimmed prior to fi rst calving (e.g. assuming 38% of herd size is replaced by heifers annually), and whether or not lame cows are seen by the trimmer (e.g. at an average rate of 40 lame cows per 100 cows per year) can easily be calculated. The recorded number of trims should be close to this expected number if in fact the herd is actually performing the hoof-care program as described. The actual recorded number of trims divided by the number of annual visits gives the esti mated number of cows trimmed per visit. The trimmer should visit the herd frequently enough to keep this number at around 50 cows per worker per visit (where workers include trimmers and assistants). Some trimmers work with multi ple chutes and have assistants to speed throughput, but the ’50 per person’ number has proven to be a useful rule of thumb. Rather than focus on the diff erent techniques for trimming, it is more important to recognise where trimming is done in such a fashion that it increases the risk of lameness. The major technique problems include: 1. Trimming the claws shorter than the recommended 75mm from coronary band to toe 2. Trimming the heel of the inner claw of the rear foot – this must not be trimmed in most cases 3. Trimming the soles too thin – so that they give under fi rm thumb pressure aft er trimming 4. Trimming the soles concave – forcing weight onto the wall and leading to spreading of the claws during weight bearing on fi rm surfaces 5. Rounding the toe off excessively – removing weight bearing wall from the toe and side wall 6. Trimming the axial wall at the toe – removing weight bearing surface and thinning the wall near the axial groove CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 124 There are other problems that can be identifi ed, such as imbalance between the inner and outer claws, the extent of modelling, poor treatment of existi ng lesions and hoof block applicati on and other issues, but the six listed above may be viewed as the most signifi cant. It is very important not to confuse lesions resulti ng from excessive wear of horn from lesions caused by excessive removal of horn – as these lesions are very similar. Most commonly we see a combinati on of excessive removal and excessive wear in large dairy herds and this presents an opportunity to educate the trimmer and adapt the technique being used. LESION ANALYSIS Hoof lesion records are notoriously variable and unreliably kept. As an aid to troubleshooti ng, I prefer to keep analysis as simple as possible, analysing by group, (default is all cows), and by whether the cow has been trimmed as part of a routi ne preventi ve program (TRIM) or because she was identifi ed with abnormal gait (LAME). Simple records of the major lesions (digital dermatiti s, foot rot, white line disease and sole ulcer) can sti ll be useful to the investi gator. Whatever the scoring system used, total counts of lesions should be obtained for each lesion type. Results are presented as a proporti on of lesions (not as a rate relati ve to a ti me period or number of cows or feet trimmed) and the aim is to focus att enti on on the dominant lesions. Pay parti cular att enti on to the diff erences between fresh cows, fi rst lactati on heifers and mature cows, the rati o of non-infecti ous lesions to infecti ous lesions and also look for trends within lesion groups. For example, the rati o of sole ulcers to white line disease (the sum of haemorrhage, fi ssure and abscess) is of interest as it leads us towards diff erent preventi ve measures (reducti on of standing ti me for sole ulcers compared to improvement of fl ooring for white line disease). Reasons for diff erences in lesion distributi on between TRIM and LAME may be apparent. For example it is common to fi nd more infecti ous subclinical lesions in the TRIM group, but more sole ulcers and white line disease in the LAME group. If we are to prevent lameness, we should focus on the lesions that dominate in the LAME records. More advanced record systems may permit an investi gati on of the days in milk at the fi rst lesion event, analysis by foot and by lactati on group and the calculati on of a true incidence rate.

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 125 INFECTIOUS HOOF LESION CONTROL We are taking an holisti c approach to infecti ous hoof lesion control, realising that there are important geneti c, nutriti onal, and environmental factors that play a role. Our data show that heifers that develop a digital dermatiti s lesion during the rearing period are at twice the risk for developing lesions once they calve compared to heifers that are disease free. This may refl ect a geneti c suscepti bility – and this is currently under investi gati on, but it may also refl ect that at least some of these animals become chronically infected ‘carrier’ cows – and become the point source of infecti on for the rest of the herd. These may be identifi ed in a herd as animals that suff er repeated bouts of digital dermatiti s infecti on and they will oft en have chronic dyskeratoti c or proliferati ve lesions oft en in associati on with severe heel horn erosion on multi ple feet. Identifi cati on of these cows and eff ecti ve treatment of these lesions is a focus of our current research eff orts, as is dietary manipulati on to reduce the severity of infecti on. Unti l then, in practi cal terms on farm, we may only signifi cantly impact two major control points – hygiene and hoofb athing. FOOT HYGIENE Poor leg hygiene is a key factor in determining risk for infecti ous hoof lesions. The greater the contaminati on, the more digital dermatiti s and interdigital phlegmon (foot rot) we observe. We score a representati ve group of cows using a simple 4-point leg hygiene scoring system (aim for at least 20% of cows in a group). The degree of contaminati on gives us a starti ng point to esti mate the required frequency of hoofb athing. Method of manure removal is also important. If the cows are in the pen when an automati c scraper pushes a wave of manure through, then leg hygiene and risk for digital dermatiti s may be very high. Removal of manure when cows are out of the pen is preferred and it is recommended that this occurs at least three ti mes per day. HOOFBATH MANAGEMENT The footbath program is a criti cal component of infecti ous hoof disease control on many farms and we may impact the effi cacy of the program through decisions related to frequency of use, design and locati on of the bath, and the type of chemical used. Leg hygiene will indicate how frequently a hoofb ath should be used. In herds where the legs and feet are clean, baths may be used on an as needed basis. In overstocked 6-row freestall barns with automati c alley scrapers, we will need to bath 5-7 days a week. Typically in most herds, the hoofb ath will be run only once a day. The footbath dimensions are criti cal for success. The bath should be at least 3.0m long to allow each rear leg to have at least 2 immersions per pass, and deep enough to allow the chemical to reach in-between the claws. We now recommend a 0.25m step-in height and fi ll the bath to 0.08- 0.10m deep. A bath width of ~0.6 m will limit the bath capacity to ~190 litres, which is the median hoofb ath capacity used on farms. Larger baths will need more chemical and be more costly to run. The bath should be located so that it minimises obstructi on to fl ow of traffi c, is in a place where it may easily be fi lled and empti ed, and where cows do not immediately enter a manure contaminated area. It should not be positi oned immediately adjacent to a wash bath – as this allows for the transfer of wash water into the treatment bath, which serves to dilute the chemical concentrati on. The fl ooring surface of the bath should not be traumati c – excessively rough or fi lled with stones and other debris, as this will inhibit the throughput of cows and potenti ally create new lesions. On days when the bath is not in use, the cows should be able to walk around the bath, not through a bath that progressively fi lls with manure during the course of the day. There is litt le data to suggest how long a chemical remains eff ecti ve, but it is prudent to change the bath aft er every 200-300 cow passes on average unless proven otherwise. Many farms will rotate pens through the bath in a diff erent order each day. We fi nd a variety of diff erent anti bacterial agents are eff ecti ve including copper sulphate, formalin and zinc products (most commonly liquid zinc chloride). All should be used at the lowest concentrati on that allows control of the disease. NON-INFECTIOUS HOOF LESION CONTROL While lesions of the claw horn are frequently blamed on nutriti on, the evidence for a true eff ect of feeding on ‘laminiti s’ in dairy catt le is scant indeed. While I fi rmly believe that cows should be fed to minimise the risk for sub-acute ruminal acidosis, and be supplemented with appropriate levels of trace minerals to opti mise immunity and horn quality, it is very rare for us to identi fy the rati on as a direct cause of lameness on a farm. In our herds, the most likely source of a nutriti onal problem is highly digesti ble maize silage that seems to cause trouble from ti me to ti me. The impact of hormonal events at calving ti me and the role of the digital fat pad appear more signifi cant in confi nement-housed dairy catt le. We know that sole ulcers behave diff erently from

white line disease, and yet we sti ll have a tendency to group these problems together. This is a mistake. White line disease in our experience may be linked to calving changes and to the impact of the fl ooring surface on the claw horn capsule. It does not appear to be related to the decrease in thickness of the digital fat pad observed at peak yield, around 120 days in milk. In contrast, sole ulcers occur early in lactati on and appear to be associated with calving and changes in the fat pad. Beyond this, sole ulcers are associated with risk factors that result in increased standing ti me per day. Based on our lesion analysis for the farm, we will emphasise some areas over others when trying to formulate a plan for improvement. Where noninfecti ous lesions dominate over infecti ous, we will focus on nutriti on and transiti on. Where sole ulcers and haemorrhage dominate white line disease, we focus on standing ti me risk factors (ti me budget, stall comfort and heat stress), and where white line disease dominates, we focus on fl ooring risk factors for slipping, trauma and wear. TRANSITION The transiti on cow assessment focuses on how heifers are transiti oned into the main lactati ng cow herd. Exposure of heifers to concrete and freestalls (cubicles) before calving is vital if they are to face this type of housing aft er calving. We recommend that heifers enter the close-up group at least 30 days before calving so that they can become accustomed to a change in diet and housing. Bedded packs provide ideal comfort but are oft en diffi cult to manage from a hygiene perspecti ve. Ideally, heifers are kept in a non-competiti ve situati on, away from mature cows throughout the transiti on period. However, if they are to be mixed at some stage during transiti on, it is bett er for this to occur as far away from calving as possible. Bunk space per cow in the close-up pen is a criti cal determinant of intake, and we recommend at least 0.76m per cow for Holsteins. In freestalls, transiti on cows and heifers should have one appropriately sized stall per cow available, or suffi cient lying space on a dirt lot or bedded pack. Transiti on in grazing herds is parti cularly traumati c to the heifer. Having been reared at grass heifers are oft en abruptly transiti oned onto the concrete of the holding area at milking ti me and join a single group of milking cows. In these herds, some att empt should be made to expose heifers to concrete prior to calving and it has proven benefi cial to manage them in a separate group aft er calving. Aft er calving, we have been experimenti ng with diff erent milking schedules. For lame cows, we CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 126 recommend no more than twice a day milking in the fresh pen, for the fi rst 21-30 days of lactati on. This may help with energy balance and give the cow more opportunity to rest. In many cases this approach may actually improve milk producti on rather than depress it! NUTRITION AND FEEDING A full nutriti onal work-up is usually beyond the scope of a lameness assessment. However, nutriti on and feeding management is an important part of lameness preventi on and we focus on feed delivery, feed access, risk factors for sub-acute ruminal acidosis and nutrient supply (with an emphasis on nutrients believed to have a role in lameness). Assessments are typically performed for close-up cows, fresh cows and early lactati on group cows. Target feed space per lactati ng cow is 0.61m. For close up cows and post-fresh cows, we prefer 0.76 m of space per cow. Cows should not return from milking to an empty bunk so if this occurs during the visit it should be noted and investi gated. TMRs that are very dry (>50%DM) are at risk of sorti ng, as are diets that are excessively long in parti cle length. The proporti on of the TMR on the top screen of the Penn-State Shaker box should be in the range 7-15%. In grazing herds, feeding of highly fermentable carbohydrate feeds in the parlour or aft er milking can be a risk factor for acidosis. For nutrient density, the paper rati on maybe consulted for inclusion rates, with emphasis on DM, fi bre, carbohydrate and trace minerals. However, it is preferable to collect a bunk sample of recently delivered fresh TMR. Representati ve subsamples of the delivered TMR should be collected throughout the enti re length of the bunk and chemical compositi on determined using wet chemistry. While there are inaccuracies in this method, it has been successfully used to identi fy gross mistakes in rati on formulati on. RISK FACTORS FOR WHITE LINE DISEASE WALKING SURFACES Walking surfaces can aff ect lameness through their ability to create trauma (an excessively rough or unfi nished surface), infl uence the tendency to slip (an excessively smooth, poorly grooved surface), create wear (either though the type of surface, its slope or distance the cow must walk upon it), and cause concussion (a parti cular problem for the lame cow walking on hard concrete). Each walking and standing surface outside the stall should be graded for each of the four risk

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 127 factors. Parti cular att enti on should be paid in situati ons where white line disease is common, and where the rati o of white line disease to sole ulcers exceeds 1:1. Problems of excessive hoof wear are observed as herds expand and we expect cows to walk more than 2 building widths (approx. 60m) to and from the milking centre. Problems are also seen when herdsmen rush the cows along uneven surfaces, making them slip. Where rubber fl ooring is used, it should be nonslip and maintained so that strips of wire from the rubber are not exposed. While benefi ts are clear in transfer lanes, holding areas and parlour fl oors, the benefi ts of rubber in front of the feed bunk in the pens have been less clear. In many herds, rubber pen fl ooring appears to reduce the risk for white line disease, but increase the risk for sole ulcerati on, without a net positi ve eff ect on overall lameness control. Results are also equivocal in the scientifi c literature, with few studies demonstrati ng a demonstrable eff ect on lameness. This could be due to the fact that with rubber fl oors and matt ress freestalls, cows appear to spend more ti me standing in the alleys on the rubber and less ti me lying down in the stall. We have worked with improved concrete fi nishes and currently prefer a grooved concrete fl oor with grooves 1.9cm wide, 1.3cm deep spaced 8.2cm on centre. RISK FACTORS FOR SOLE ULCERATION TIME BUDGET Daily lying ti me is a very important factor predisposing catt le to lameness – especially due to sole ulcer and sole fracture/heel ulcer. Standing and lying ti mes are aff ected by many diff erent factors that impact the daily ti me budget. In confi nement freestall housing and dirt lot herds fed a Total Mixed Rati on (TMR), feeding ti me is typically 4.5 hours per day (h/d). For grazing herds, feeding ti me is esti mated at 8h/d. Drinking ti me is esti mated at 0.5h/d (including the ti me spent near the trough). Time out of the pen milking is the most common way we negati vely impact the ti me available for rest in dairy herds and it may be esti mated from parlour throughput or accurately ti med for the high producti on group. In grazing and dirt lot dairies, ti me traveling to and from the parlour and ti me held off paddocks not milking must be taken into considerati on. In confi nement herds, the remainder of the ti me in the pen is split between ti me in the stall and ti me in the alley. Typically, cows spend 2h/d socialising in the alley in confi nement freestalls and in the paddocks in dirt lot dairies. In an overstocked pen, lying ti me will be limited by access to a stall once stocking density exceeds 1.2 cows per stall. Once in a stall, the ti mes spent standing and lying in a confi nement freestall herd are determined by lameness. Lame cows stand for longer in the stall and this may be esti mated using an index of comfort called the ‘Stall Standing Index’. This must be measured approximately 2 hours before the morning or aft ernoon milking (morning is preferred in hot climates) and the index is the proporti on of cows touching a stall (lying + standing all the way in the stall + standing half in and half out of the stall) that are standing (standing all the way in the stall + standing half in and half out of the stall). For example, if 80 cows are touching a stall and 20 cows are standing in a stall, the SSI is 0.25 or 25%. The target is 20% or less. An equati on uses this index to esti mate the ti me standing in the stall. Time available for lying is equal to 24h minus the ti me spent performing each of the acti viti es described above. The target is a minimum of 12h/d for mature cows. LYING COMFORT Bedded packs or loose housing provide a comfortable lying area for the cow and they are oft en used for the peri-parturient cow. There are two main types of bedded pack, uti lising either aerobic (compost) or anaerobic fermentati on. Whatever the type, it is important to provide suffi cient bedded area per cow and this is calculated based on an esti mate of the average body weight in the group. For aerobic packs, the bed is allowed to accumulate over many months, but the bedding must be sti rred at least twice a day. Fresh bedding is added as required to keep the cows clean and dry. For anaerobic packs, the bedding must be completely removed every 4-5 weeks and fresh bedding added daily at a rate of approximately 11 kg per cow per day. All packs should be demarcated from the concrete feeding alley by a bedding retainer, used to hold the bedding back. The concrete feed alley should be scraped at least twice a day and the area should be well venti lated with the use of recirculati on fans over the pack. Drainage is important to reduce humidity in the barn, and cows must not be able to access water troughs from the bedded area – only from the concrete feed alley side, otherwise large areas of bedded area will become water logged and contaminated. Where bedding material is less available, freestalls are more commonly used for housing dairy catt le.

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 128 The freestall is assessed from the perspecti ve of the cow entering, lying down, standing and leaving the stall. To that end, we focus on fi ve key aspects of design: surface cushion and tracti on; resti ng area and how it is defi ned in front by the brisket locator and to the sides by the divider loops; room to lunge (horizontal movement of the head) and bob (verti cal movement of the head) when rising and lying down; space below and behind the neck rail to stand and rise unhindered; and a rear curb low enough to facilitate exit from the stall. Stall surface opti ons commonly used include deep loose bedding with sand, sawdust, and manure solids or a matt ress – typically made of rubber crumbs, foam or water. It is important to make a disti ncti on between the interacti on of lame and non-lame cows with the surface. Sand and deep bedded compost will facilitate the rising and lying movements of lame cows more than mats or matt resses, while matt resses will perform quite well for non-lame cows. For lameness control, sand is the gold standard bedding material we recommend. HEAT ABATEMENT Sole haemorrhage and ulcerati on commonly occurs in North America at the end of the summer, early fall – typically 2 months aft er the peak in heat stress. Cows stand ~3h longer under conditi ons of mild to moderate heat stress and we believe that this is a major reason for the increase in lameness at this ti me. Venti lati on is criti cally important during the hot summer months to keep cows cool and limit changes in behaviour and feeding acti vity. Opti ons for venti lati on are natural or tunnel/ mechanical venti lati on. Naturally venti lated barns are preferably orientated east to west in North America, to maximise capture of summer cross winds. Eaves should be at least 3.7m high, with a minimum 4:1 roof pitch. Ridge opening should be 0.05m per 3m of building width and it should be possible to open more than 50% of the side wall. Mechanically venti lated tunnel barns require suffi cient fan capacity at one end of the barn to draw air from the opposite end at a rate of 1.1m per sec. To venti late the barn during the summer at 40-60 air changes per hour, we must check that we have enough fans and that the inlet area is sized appropriately. The keys to heat abatement other than barn venti lati on are water availability, suffi cient fans located over the resti ng and standing areas to move air over the cows, use of soaking to wet the cow to facilitate cooling by evaporati on, or misti ng to cool the air moving over the cow, and the provision of enough shade. Targets for cows in hot climates are to provide at least 9cm of water trough perimeter per cow, to have appropriately sized and spaced fans over the resti ng area, set to acti vate at a temperature of 21o C, and to have an eff ecti ve soaking system that soaks cows long enough, frequently enough, and at the right temperatures. The holding area is a criti cal place for lameness control, infl uencing ti me out of the pen, foot trauma, and heat stress. It is generally accepted that we must minimise ti me out of the pen for milking to less than 3h/d. Ideally cows should spend a maximum of 45 minutes to one hour per milking out of the pen. The fl ooring surface and the use of the crowd gate should minimise the chances of traumati sing the hoof. Minimum space requirement for the holding area is suggested at 1.8 square m per cow. For cooling, it is recommended that we provide at least 28 cubic m per min per cow air movement from the recirculati on fans, and that soaking is aggressively applied at a temperature of 21o C. CONCLUSION The complex multi factorial causality of lameness hinders our ability to communicate simple improvements to management and housing that herd owners can implement successfully to achieve dramati c reducti ons in lameness. It does not have to be that way. The troubleshooti ng template described has been successfully used on many farms world-wide under diff erent management systems to improve lameness control – for the good of the cows and for the good of the herd owner.

Key Performance Indicators for the UK nati onal dairy herd obtained from milk recording data Hanks, J.D.1 , Kossaibati , M.A.2 , 1 Veterinary Epidemiology & Economics Research Unit (VEERU), SAPD, University of Reading, P.O.Box 237, Reading, RG6 6AR 2 225 Cassiobury Drive, Watf ord, WD17 3AN ABSTRACT A sample of 500 herds was selected at random from all Holstein-Friesian dairy herds on monthly assisted milk recording with Nati onal Milk Records (NMR). A series of ferti lity, producti on and health parameters were calculated for each herd covering the 12 month period to the end of September 2010. Each parameter was analysed separately to identi fy the median and inter-quarti le range values across the 500 herds. The practi cal use of these parameters is discussed with regard to highlighti ng herd strengths and weaknesses as well as sti mulati ng discussion and collaborati on between farmers and their technical advisers. INTRODUCTION Dairy producti on is a dynamic process where sustainable and profi table output is dependent on good performance in many diff erent aspects of herd management, health and welfare. In recent years there has been much greater use of Key Performance Indicators (KPIs) to benchmark performance and highlight areas of concern at the earliest opportunity (Hanks 2007). While it is relati vely straightf orward to calculate producti on parameters for an individual herd, this alone does not answer the fundamental questi ons of “What is a good or acceptable level of performance in a commercial herd?” and “Where might an individual herd improve?” Answering these requires comparison of herd performance against a large representati ve sample of other herds. MATERIALS AND METHODS A sample of 500 commercial dairy herds was selected at random from all predominantly Holstein-Friesian herds that milk record on a monthly basis with Nati onal Milk Records (NMR) in the United Kingdom. All herds used fully assisted recording, thus excluding DIY recording herds. To remove the infl uence of seasonal fl uctuati ons in performance, 12 month rolling average fi gures were calculated using the InterHerd+ program, for a total of 63 parameters, describing performance for the year completed on 30th September 2010. For each parameter the values for all 500 herds CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 129 Figure 1. A descripti on of the median, inter-quarti le range and target values generated for each parameter.

were arranged in ascending order. A median value and inter-quarti le range were derived for each parameter. The inter-quarti le range was used to identi fy a “Target” value for each parameter, based on the performance level achieved or bett ered by 25% of herds. Thus for parameters like concepti on rate and 305 day yield the target would be the higher value of the inter-quarti le range. For other parameters like somati c cell count and culling rate the target would be the lower value of the interquarti le range (see Figure 1). CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 130 RESULTS The median, inter-quarti le range and target values for 25 key parameters are given in Table 1. DISCUSSION This study was originally undertaken to identi fy realisti c target and interventi on values for use in the InterHerd+ program. The combinati on of such a large sample of herds and the use of the interquarti le range to generate an objecti ve measure of target values has resolved an area of frequent argument. These nati onal stati sti cs have helped Parameter Median 1st-3rd quarti le (25% - 75%) Target (“best” 25%) Inter-quarti le range A. Culling rate 24% 18% - 30% 18% 12% B. Culling/death rate in fi rst 100 days of lactati on 7% 4% - 10% 4% 6% C. Percentage Served by day 80 46% 32% - 59% 59% 27% D. Percentage conceived 100 days aft er calving 26% 16% - 33% 33% 17% E. Calving to 1st service interval (days) 105 87 – 124 87 37 F. Calving interval (days) 424 409 – 441 409 32 G. Age at 1st calving (years) 2.4 2.3 - 2.7 2.3 0.4 H. Concepti on rate 32% 26% - 40% 40% 14% I. Percentage service intervals at 18-24 days 30% 22% - 38% 38% 16% J. Percentage service intervals >50 days 32% 22% - 43% 22% 21% K. Percentage eligible for service that were served 27% 19% - 37% 37% 18% L. Percentage eligible for service that conceived 9% 5% - 13% 13% 8% M. Lifeti me milk/cow/day (kg) 11 8 – 13 13 5 N. Milk/cow/day (kg) 21 18 – 24 24 6 O. Average protein% 3.27% 3.19% - 3.33% 3.33% 0.14% P. Average fat% 3.96% 3.82% - 4.12% 4.12% 0.26% Q. 305 day yield (kg) 7,400 6,500 - 8,300 8,300 1,800 R. Average SCC (‘000 cells/ml) 210 169 - 268 169 99 S. Percentage SCC >=200,000 cells/ml 24% 19% - 31% 19% 12% T. Percentage SCC >500,000 cells/ml 9% 7% - 12% 7% 5% U. Percentage 1st recording SCC >=200,000 cells/ml 20% 15% - 25% 15% 10% V. Percentage chronic SCC >=200,000 cells/ml 14% 10% - 19% 10% 9% W. Percentage Dry period cure (High:Low) 74% 64% - 80% 80% 16% X. Percentage Dry period protecti on (Low:Low) 84% 77% - 89% 89% 12% Y. Percentage Low at end of previous lactati on (SCC<200,000 cells/ml) 60% 49% - 70% 70% 21% Table 1. Summary of Key Performance Indicators for the year ending 30th September 2010 derived from analysis of 500 Holstein/Friesian milk recording herds.

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 131 farmers and advisers to identi fy quickly the areas of greatest concern and focus discussions more on the causes and possible soluti ons to poor performance. The wide variati on that exists between the best and worst performing herds is evident from Table 1, as well as graphs of the distributi on for individual parameters. In additi on to suggesti ng considerable scope for improvement in the majority of herds, these fi gures also provide a means for identi fying “best practi ce”. It is equally important to know, and learn from, the farmers where performance is outstanding (Hanks & Kossaibati 2010). The Key Performance Indicators Report in the InterHerd+ program uses the values derived by this study to compare the performance of an individual herd against these nati onal target values (Figure 2). The report displays a number of Key Performance Indicators (KPIs) that cover the producti on, ferti lity and health status of an individual herd. The herd’s current 12 month rolling average performance for each parameter is displayed as a value to the left of each parameter ti tle. The Target and inter-quarti le range values are also displayed to the right of the parameter ti tle. The Target and Range values are also used to display the performance graphically for the herd. In the coloured graphic to the left of the report the herd’s 12 month performance for each parameter is displayed as a black square („). The verti cal black line represents the Target value. To the left of the target line represents performance worse than target while to the right of the target line indicates performance bett er than target. Thus, where the black square is to the right of the Target line, the herd would have appeared in the top 25% of herds for that parameter. The verti cal line to the left of the target line represents the positi on that would correspond to the target value worse by the inter-quarti le range. Hence for any parameter where the black square is to the left of the line then that herd would appear in the bott om 25% of herds when compared to the survey of 500 herds. The arrows indicates the performance level for the latest 3 months. This can provide an indicati on of the directi on of movement in a parameter, but must be interpreted with cauti on as it may indicate a response to seasonal or other short-term factors. The Key Performance Indicators Report is an Figure 2. Key Performance Indicators Report comparing performance of a herd against the target and range values from the study.

ideal document for sti mulati ng discussion between the client and their technical advisers. While there may be enti rely logical reasons for some of the high or low values (reducing herd size would generate a high culling rate for example), this report will also highlight areas where performance could be improved. Comparing performance against a large group of commercial herds is an uncontroversial and eff ecti ve means of engaging farmers in considering the underlying causes of poor performance. As an example, the herd displayed in Figure 2 has poor somati c cell count performance, parti cularly relati ng to dry period protecti on. Similarly, the parameters relati ng to inter-service intervals strongly suggest fundamental heat detecti on problems and too many PD-ves. Similar studies have also been conducted by the authors for herds that are predominantly Jersey, Guernsey and Ayrshire (Hanks & Kossaibati 2011a, b and c respecti vely). Similar studies are underway relati ng to the performance of organic herds. ACKNOWLEDGEMENTS The authors are very grateful to Nati onal Milk Records (NMR) for permission to use the milk recording data for this analysis. CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 132 CONFLICT OF INTEREST DECLARATION Dr James Hanks is also a director of PAN Livestock Services Limited. This company owns and develops the InterHerd+ program. REFERENCES Hanks, J.D. (2007) What performance should a herd achieve? Comparing the strengths and weaknesses of your clients’ herds. UK Vet 12, No 1 Hanks, J.D., Kossaibati , M.A. (2010) A study of herd performance in 500 milk recording herds. htt p://www.veeru. reading.ac.uk/secti on1/Research.htm Accessed August 2011 Hanks, J.D., Kossaibati , M.A. (2011a) Jersey herds: The performance across 83 Jersey herds for the year ending 31/12/2010 htt p://www.veeru.reading.ac.uk/secti on1/ Research.htm Accessed August 2011 Hanks, J.D., Kossaibati , M.A. (2011b) Guernsey herds: The performance across 33 Guernsey herds for the year ending 31/12/2010 htt p://www.veeru.reading.ac.uk/secti on1/ Research.htm Accessed August 2011 Hanks, J.D., Kossaibati , M.A. (2011c) Ayrshire herds: The performance across 47 Ayrshire herds for the year ending 31/12/2010 htt p://www.veeru.reading.ac.uk/secti on1/ Research.htm Accessed August 2011

How to determine on farm forage mineral problems Bone, P.A.1 , Kendall, N.R.2 , 1 39 Stratt on Heights, Cirencester, GL7 2RH 2 Lecturer in Nutriti on, School of Veterinary Medicine and Science, University of Notti ngham, Sutt on Bonington Campus, Loughborough, Leicestershire, LE12 5RD ABSTRACT Forages harvested or grazed at diff erent locati ons may have drasti cally diff erent mineral profi les. This may lead to imbalances of many minerals giving rise to clinical conditi ons or ineffi ciencies in producti on. The aim of this paper is to outline sources of informati on and how to interpret them to enable veterinarians to prevent, diagnose and/or treat the clinical conditi ons/ineffi ciencies associated with mineral imbalances. Forage mineral profi les, analysis of drinking water and total mixed rati on mineral profi les are key determinates to assess on farm minerals. Book values or manufacturers’ declarati ons of mineral levels can be used for compounded feed and other feedstuff s unless fed a large amount from one (usually home-grown) source, when analysis would be recommended. Total inputs should be calculated for each mineral. Mineral interacti ons should be considered. This should allow an esti mati on of any potenti al issues, which may allow appropriate supplementati on/management changes to be undertaken to prevent occurrence. However, to diagnose a clinical problem analysis of the appropriate sample from the animal should be done. KEYWORDS: Forage, silage, mineral, trace element, diagnosis INTRODUCTION Forages have mineral compositi ons largely determined by the local geology, which can vary widely even within the same fi eld. Therefore the same forages harvested or grazed at diff erent locati ons and even the same locati on over diff erent ti me-points may have drasti cally diff erent mineral profi les. This may lead to imbalances of many minerals giving rise to clinical conditi ons or ineffi ciencies in producti on. The major eff ects on producti on and ferti lity can be considered to be, in potenti al order of importance: management; disease; energy; protein; water; forage and soil (Figure 1). Minerals have a role to a greater or lesser extent within each of those categories, for example disease incidence can be aff ected by immune status and many minerals have major immunological roles. We aim to outline the sources of informati on and enable veterinarians to interpret and uti lise these to aid on farm diagnosis of mineral conditi ons. BASIC MINERAL NUTRITION There are 15 minerals which are essenti al nutrients. These can be broken down into two groups: macro and micro minerals (Table 1 and 2). All mineral elements are interlinked which can make it diffi cult to determine nutriti onal status e.g. CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 133 Figure 1. A representati on of what aff ects animal health and ferti lity in the dairy cow. Table 1. 7 Macro elements Functi on Potassium* (K) Water balance* Calcium (Ca) Bone Chloride (Cl) Water balance Phosphorus (P) Energy Magnesium (Mg) Bone & metabolism Sodium (Na) Water balance Sulphur* (S) Protein & skin * Table 2. 8 Micro elements Functi on Iron* (Fe) Blood, Energy* Zinc (Zn) Feet, skin, udder Manganese (Mn) Health, Ferti lity Copper (Cu) Growth, health, energy, ferti lity Iodine (I) Energy, ferti lity Cobalt (Co) Energy, health, growth Selenium (Se) Health, ferti lity, udder, growth Molybdenum* (Mo) Protein* Those elements marked * are of major interest and will be dealt with later in the paper.

defi ciency, toxicity, suffi ciency, excess. Someti mes an excess of an element can manifest similar clinical signs to the defi ciency. Therefore it is very important to have the correct informati on about input levels of macro and micro elements to the animal. To determine mineral problems one should take into account the total rati on consumed. This should consider: • All feed inputs. owParlour feed. Plus full mineral specifi cati ons. owBlend. Plus full mineral specifi cati on or background levels. owStraights. Background mineral levels. owAll forages. Mineral levels. owGrazing. Mineral levels. • Water. Only if private supply or hard water areas. • Supplemental mineral inputs. This must include bagged mineral, buckets, injectable minerals, drenches & boluses. • Soil. Full soil analysis, if needed. This is not necessarily the formulated rati on as rati on fed can be diff erent and the rati on actually consumed be even more diff erent due to selecti on within the diet. Especially re the amounts of free access mineral consumed. Table 3 shows the various mineral inputs and sources of informati on. This informati on can then be entered into a computer based calculati on system which are stand alone or within rati on programs. This will allow the user to see supply against need and determine what further steps/ analysis is required. The informati on on mineral content of rati on components can be gained relati vely easily. Sources of this informati on are outlined in Table 3. This will then allow the user to guide their clients to the next stage of the investi gati on process. Mineral inputs can be entered to a mineral rati on CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 134 calculator program. Mineral check is an example of one of these mineral calculators. These programs allow the user to ascertain the amount of mineral elements being fed to the livestock group under investi gati on. The soft ware allows the user to view all the mineral inputs and enables a judgement to be made on what other investi gati ons can be undertaken. An example of a typical output is displayed in Figures 2a & 2b. Figures 2a and 2b show the typical output from the Mineral Check Program. Screen (a) shows the background mineral inputs whilst screen (b) shows the contributi ons from background and additi onal supplements and relates this to target levels to specify balancer minerals required. Note that all targets and inputs can be changed. In the example shown in Figures 2a and 2b, the user should be able to see that the in-put for calcium within the post calving diet is low against the requirement therefore to increase the calcium within the dry matt er intake the user could consider recommending the use of limestone fl our at 120gms/milking cow per day fed within the TMR diet. This level of limestone fl our will give 50gms of calcium per milking cow per day. The use of blood for assessment of status of macro elements does not produce indicati ve results as the ruminant animal homeostati cally holds blood levels of calcium in a very ti ght band. Urine levels can provide a bett er picture of adequacy of intake and this is true for most macro minerals including magnesium and phosphorus. TRACE ELEMENT INVESTIGATIONS Investi gati ons for trace element defi ciencies are more complex to deal with. The program allows an insight into the inputs against the published requirements of the stock type being dealt with. There are a number of stages to consider: the program does not fully consider all mineral interacti on and does not diff erenti ate legal Table 3. Parlour feed, blends & straights From feed manufactures and suppliers Forage, grazing & livestock drinking water mineral reports & soil Available from a number of laboratories. (See notes) All other mineral inputs This must include bagged mineral, buckets, injectable minerals, drenches & boluses This informati on can be found on product’s specifi cati on or manufactures web sites. The most important informati on is feeding rate or supply of mineral elements per day.

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 135 Figures 2a & 2b. a) b)

maximums from the published requirement. The legal maximum is higher than published requirements for all elements and the legal maximum should never be considered to be the formulatory requirement when calculati ng rati ons. For example the stated daily requirement for copper for a dairy cow (ARC 1980) is 11mg/kg Dry Matt er (DM), however the legal maximum is 35mg/ kg which equates to approximately 40mg/kg DM. Within the program a value of 20mg/kg/DM has been used following the recommendati ons of the Defra copper guidance note (Defra 2011). For trace element status diagnosis within the animal, the approach depends upon the element(s) under considerati on and the status being investi gated. For suspected high copper status then it is best to analyse liver copper concentrati ons from contemporary cull animals or via biopsy. Liver copper is not as useful for diagnosis of copper responsive conditi ons. For selenium plasma selenium and erythrocyte glutathione peroxidise (eGSHPx) will both give informati on about dietary intake and functi onal use respecti vely, as with many diff erent analyses they look at status over diff erent ti me periods, a low eGSHPx and high plasma selenium may be indicati ve that selenium supplementati on has commenced a week or two prior to sampling from an animal in a previously defi cient situati on. CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 136 CONCLUSION When evaluati ng mineral status on farm then we must consider total dietary inputs including supplements and water. This can provide investi gati ve leads to inform within-animal tests. Computer rati oning programs can be useful but are dependent on the data inputt ed, both in terms of requirements and dietary intake. Many of the (sub)clinical signs of mineral conditi ons can be caused by other factors including husbandry, general nutriti on and environment. ACKNOWLEDGEMENTS Dr David Atherton, Thomson and Joseph Ltd for access to Mineral Check Program. REFERENCES ARC (1980) The Nutrient Requirements of Ruminant Livestock, Commonwealth. Agricultural Bureaux, Slough. Defra (2011) htt p://www.vla.defra.gov.uk/reports/docs/ rep_copper_guidance.pdf accessed 23-8-11.

The environmental impact of dairy and beef producti on: improving producti vity off ers miti gati on opportuniti es Capper, J.L., Assistant Professor of Dairy Science, Department of Animal Sciences, Washington State University, Pullman, Washington, 99164-6310, USA ABSTRACT A greater quanti ty of animal protein will be required to feed the future global populati on in an environmentally sustainable manner, thus livestock producers are faced by the challenge of producing more milk or meat using fewer resources (land, water and energy), with a smaller carbon footprint. Improved producti vity has considerably reduced the carbon footprint of dairy and beef producti on over the past century, yet the consumer oft en mistakenly perceives extensive systems to be intuiti vely more environmentally-friendly than their intensive counterparts. This paper will examine the quanti tati ve eff ects of improved producti vity in the dairy and beef industries on total environmental impact per unit of food produced. KEYWORDS: Carbon footprint, environmental impact, beef, dairy, producti vity, diluti on of maintenance INTRODUCTION In 1800, each U.S. farm could only produce enough food to feed one extra family. In the wake of considerable improvements in effi ciency and producti vity, each farmer currently produces, on average, enough food to feed 125 other people. However, we remain in the midst of a global food crisis, with the number of food-insecure people in the world increasing from 820 million in 2004- 2006 to 1,020 million in 2009 (Food and Agriculture Organizati on of the United Nati ons 2009b). The global populati on is predicted to increase to over 9 billion people in the year 2050 (U.S. Census Bureau 2008). Projecti ons indicate that the average domesti c income will also increase, with the projected GDP of China and India being similar to that of the United States (Keyzer and others 2005). The desire for a diet richer in animal-source proteins rises in tandem with increasing income, thus the global livestock sector will be charged with the challenge of producing more milk, meat and eggs using fewer resources. To fulfi ll total food requirements, the Food and Agriculture Organizati on of the United Nati ons (FAO) predict that 70% more food will have to be produced (Food and Agriculture Organizati on of the United Nati ons 2009a), yet the area of arable land available will decrease from 0.5 ha/capita in 1950 to approximately 0.15 ha/capita in 2050 (Bauman & Capper 2011). It is clear that producti vity (i.e. food output per unit of resource input) will have to increase considerably in order to support the rise in populati on. If the present competiti on for energy, land and water supplies conti nues, resources available for agricultural producti on are likely to decrease concurrently with increased populati on growth. The global livestock industries therefore face the challenge of producing suffi cient nutriti ous, safe, aff ordable animal protein to meet consumer demand, using a fi nite resource base. ENVIRONMENTAL IMPACT AND THE “DILUTIONoOFoMAINTENANCE”oEFFECT In any agricultural or industrial sector, improving producti vity allows system fi xed costs to be diluted out over greater units of producti on, thus reducing the economic cost per unit. The same concept can be applied to environmental costs with carbon, the fundamental unit of energy in living systems, as the currency. All animals require a daily quanti ty of nutrients to maintain vital functi ons and minimum acti viti es (maintenance requirement). This can be considered to be the “fi xed cost” of livestock producti on that must be met for every animal within the populati on. The level of animal protein producti on (milk yield, growth) then determines the additi onal “variable costs” of the system. The daily populati on nutrient requirement is a valid proxy for natural resource use (land, water, fossil fuels) and greenhouse gas (GHG) emissions. Management practi ces that improve effi ciency, reduce the mass of the animal populati on required to produce a set quanti ty of animal protein and thus decrease the total associated maintenance nutrient requirement, will therefore reduce resource use and waste output per unit of food produced. Effi ciency improvements made by the U.S. dairy industry over the past 60 years may be used as a proof of this concept. As described by Capper and others (2009b), the U.S. dairy herd peaked in 1944 at 25.6 million cows, with a total milk yield of 53.0 billion kg produced through an extensive pasture-based system. By comparison, the 2007 dairy herd contained 9.2 million cows producing 84.2 billion kg milk, the four-fold increase in milk yield per cow and consequent improvement in CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 137

effi ciency being facilitated by improvements in management, nutriti on and geneti cs. Analysing the environmental impact of the US dairy industry in 2007 revealed that compared to 1944, only 21% of the dairy populati on (lactati ng cows, dry cows, heifers and bulls), 23% of the feedstuff s, 10% of the land and 35% of the water were required, and only 24% of the manure was produced per unit of milk. In consequence, the total GHG emissions (carbon footprint) per unit of milk were reduced by 63%, and the carbon footprint of total dairy producti on (up to the farm gate) was 41% lower in 2007 compared to 1944 (Figure 1). REGIONAL VARIATION IN CARBON EMISSIONS PER UNIT OF MILK If we examine internati onal trends, increased milk producti on has a miti gati ng eff ect upon carbon emissions on a global basis. The trend for producti vity to improve over the past decades is not exclusive to the U.S.; major milk-producing regions (e.g. U.S, Canada, New Zealand and Europe) have all improved milk yield per cow since the 1960s, the rate of improvement varying from 129kg/year and 117kg/year for the U.S. and Canada respecti vely, to 77kg/year and 24kg/year for Europe and New Zealand (Capper and others 2009a). The environmental eff ects of regional variati ons in producti vity are exemplifi ed by the results of a recent FAO (2010) report that modelled GHG emissions from dairy producti on using life cycle analysis. As intensity of producti on declined and the average milk yield shift ed from approximately 9,000kg/cow for North America to ~250kg/cow for CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 138 Sub-Saharan Africa, the carbon footprint increased from 1.3kg CO2 -eq/kg milk to 7.6kg CO2 -eq/kg milk. The challenge of producing more animal protein to fulfi ll human populati on requirements while minimising resource use and waste output is not confi ned to future populati on scenarios. In 2007, the Chinese government announced that the human recommended daily intake (RDI) of dairy products should be increased from 100g to 300g. Given the size of the Chinese populati on, this would require an additi onal 65 million dairy animals at current daily milk yields (11kg/d). If producti vity was improved to that of the average U.S. dairy cow (29kg/d), this would sti ll require an increase in the dairy populati on, yet this increase would be confi ned to 23 million animals with concurrent comparati ve reducti ons in total maintenance requirements, resource use and GHG output. Sustainability is oft en defi ned as having three interrelated components: environment, economic and social, with sustainability occurring through a balance of these factors. When assessing the sustainability of dairy systems the questi on should not be limited to the environmental impact of the dairying within a specifi c region, but must also consider the economic and social implicati ons. While the FAO (2010) data could provoke the conclusion that all regions should adopt North American and Western European-style producti on systems, or that dairying should be focused in these areas and be discouraged in less producti ve regions such as Sub-Saharan Africa and South Asia, the signifi cant social (both status and nutriti onal) and economic value of dairying in lessFigure 1. Resource inputs and waste output associated with dairy producti on in 2007 compared to 1944 (adapted from Capper and others 2009b).

CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 139 developed regions must not be underesti mated. The challenge for global dairy producti on is to opti mise sustainability within each region rather than prescribing the best “one-size-fi ts-all” global system. ORGANIC SYSTEMS HAVE A GREATER ENVIRONMENTALoIMPACT As seen in many global regions, milk yields from conventi onal U.S. dairy herds are signifi cantly higher (10,062kg/yr) compared to yields from organic (7,425kg/yr) or grazing herds (7,213kg/ yr; USDA 2007). This decline in producti vity has a signifi cant eff ect upon resource use. Capper and others (2008) modelled the eff ect of supplying the projected U.S. populati on in 2040 with 0.71 litres of milk (or its low-fat equivalent) per day as recommended by USDA (2005). Assuming that current producti vity trends conti nue for both crop and animal producti on, fulfi lling dairy requirements via organic producti on would increase the nati onal herd size by 3.5 million animals (20%) compared to conventi onal producti on, and augment land requirements by 3.1 million ha (a 30% increase). The world record for dairy producti on is currently held by a Wisconsin dairy cow that produced 32,726kg of milk over 365 days in 2010. Given that the average U.S. cow produced 9,593 kg of milk in 2010 (USDA 2011), considerable progress can conti nue to be made in order to improve producti vity and reduce environmental impact. EFFICENCY IMPROVEMENTS OVER TIME IMPROVEoBEEF’SoCARBONoFOOTPRINT A similar patt ern of correlated improvement between producti vity and carbon emissions has been demonstrated by the U.S. beef industry, with average beef-carcase yield per animal increasing from 274kg in 1977 to 351kg in 2007 (USDA 1978; USDA/NASS 2008). This increase in yield allowed the U.S. beef industry to produce an approximately equivalent amount of beef from four animals in 2007 vs. fi ve animals in 1977. Management advances over this ti me period also facilitated an increase in growth rate, reducing the total days from birth to slaughter and thus the total maintenance resource cost and waste output associated with beef producti on. Between 1977 and 2007, increasing growth rate meant that the age at slaughter was reduced from 609 days to 485 days. In combinati on with the increased beef yield per animal, this reduced animal numbers by 30%, feed use by 19%, water use by 12%, land use by 33%, manure producti on by 18% and the carbon footprint per unit of beef by 16% (Capper 2011). Yield metrics for meat producti on relate to the quanti ty of edible protein produced per animal, i.e. the combinati on of slaughter weight and the proporti on of the carcase that is meat vs. nonedible by-products. Anecdotal evidence from the U.S. beef processing industry indicates that beef animal slaughter weight reached a plateau and that slaughter weight cannot conti nue to increase without reorganisati on of the processing infrastructure, currently designed for an upper threshold of approximately 635kg (average U.S. beef slaughter weight for 2010 was ~590kg). The proporti on of the gains made between 1977 and 2007 that can be att ributed to improved beef yield may not be further realised in future years. Nevertheless, the beef industry has a considerable opportunity to improve producti vity through increased growth rate and lean muscle accreti on. THE MYTH OF INHERENTLY ENVIRONMENTALLY SUSTAINABLE GRASS-FINISHED BEEF The FAO (2006) concludes that it is essenti al to conti nue to intensify livestock producti on in order to maintain the effi ciency gains that improve environmental sustainability. By contrast, consumers oft en assume that pasture-based or organic beef systems are inherently more environmentally-friendly than conventi onal corn-based fi nishing systems. Growth rates are considerably lower in animals fi nished on grass and it is diffi cult to achieve high slaughter weights, therefore grass-fi nished catt le are usually slaughtered at around 486kg at 679 days of age, compared to 569kg at 453 days of age in a conventi onal system consisti ng of seven to 12 months on pasture before intensive corn-based fi nishing in the feedlot (Capper 2010). Within conventi onal U.S. systems, producti vity-enhancing technologies such as ionophores, steroid hormone implants, in-feed hormones and beta-agonists have been widely adopted (USDA 2000). However these technologies are not permitt ed within organic producti on, leading to effi ciency losses. Fernàndez and Woodward (1999) compared performance parameters for beef animals fi nished in organic or conventi onal feedlot systems and reported decreases in growth rate and feed effi ciency (1.40kg/d and 7.57kg feed per kg gain for the organic system, 1.77kg/d and 5.44kg feed per kg gain for the conventi onal system), leading to a reduced slaughter weight (536kg vs. 578kg), increased days in the feedlot (226d vs. 164d) and an increase in total producti on costs of $0.51 per kg gain ($1.86/kg gain vs. $1.35/kg gain), a cost which would ulti mately be passed to the consumer. This

comparison is somewhat disingenuous, as feedlot fi nishing is not routi nely practi sed within organic producti on – grass-fed fi nishing systems (without the use of producti vity-enhancing technologies) are far more common. As a consequence of the reduced nutrient density of forage-based diets, producti vity indices in grass-fed systems are reduced, with growth rates averaging 0.59kg/d over the enti re lifespan compared to 1.74kg/d in a conventi onal system (Capper 2010). Capper (2010) analysed these producti on diff erences using a whole system approach, incorporati ng all resource inputs and waste outputs from beef producti on with boundaries set initi ally at the manufacture of cropping inputs and terminati ng at the slaughterhouse door. If the total U.S. beef produced in 2010 (11.8 billion kg) was produced from a grass-fi nished system, an extra 64.6 million animals would need to be added to the nati onal herd, an extra 53.1 million ha of land would be required (equal to 2.1 x the land area of the United Kingdom), and the extra water required would be suffi cient to supply 79.5 million UK households for a year (adapted from Capper 2010). Despite the popular percepti on that extensive systems are more environmentally-friendly, the increase in GHG emissions produced by changing to a grassfed system would be equal to adding 26.6 million cars to the road per year. There can be no doubt that improving milk producti on, slaughter weight and growth rate reduce both resource input and waste output in dairy and beef catt le respecti vely. Nonetheless, the impact of further producti vity measures relati ng to reproducti on, health and welfare are notably lacking from current scientifi c literature and warrant further investi gati on to facilitate recommendati ons for conti nuous future improvement. Furthermore, the downstream environmental impacts of dairy and beef processing, distributi on and consumpti on are integral parts of animal protein producti on and must be assessed in order to view the full picture relati ng to livestock system sustainability. CONCLUSION The livestock industry faces a considerable challenge in producing suffi cient animal protein to feed the growing populati on while conti nuing to improve sustainability. As demonstrated by examples of improved effi ciency in the U.S. livestock industry, one soluti on is to implement producti vity gains that reduce resource use and cut carbon emissions from livestock producti on. However, it is important to note that producti on systems vary widely between regions and no specifi c “one-size-fi ts-all” practi ce CATTLE PRACTICE VOLUME 19 PART 2 BCVA 2011 140 may be recommended that will balance resource availability and beef demand. Rather than focusing on one single metric, true sustainability within the livestock industry can only be achieved by a regionspecifi c balance between environmental impact, economic viability and social responsibility. REFERENCES Bauman, D.E., Capper, J.L. (2011) Future Challenges and Opportuniti es in Animal Nutriti on. 26th Southwest Nutriti on & Management Conference, Tempe, AZ, February 24: 2011 Capper, J.L. (2010) The environmental impact of conventi onal, natural and grass-fed beef producti on systems. Proc. Greenhouse Gases and Anim. Ag. Conf. 2010, Banff , Canada, October 3-8, 2010 Capper, J.L. (2011) The environmental impact of U.S. beef producti on: 1977 compared with 2007. J. Anim. Sci. In press Capper, J.L., Cady, R.A., Bauman, D.E. (2009a) Demysti fying the environmental sustainability of food producti on. Proc. Cornell Nutriti on Conf., Syracuse, NY, 20-22 October 2009 Capper, J.L., Cady, R.A., Bauman, D.E. (2009b) The environmental impact of dairy producti on: 1944 compared with 2007. J. Anim. Sci. 87: 2160-2167 Capper, J.L., Castañeda-Guti érrez, E., Cady, R.A., Bauman, D.E. (2008) The environmental impact of recombinant bovine somatotropin (rbST) use in dairy producti on. Proc. Natl. Acad. Sci. U. S. A. 105: 9668-9673 Fernàndez, M.I., Woodward, B.W. (1999) Comparison of conventi onal and organic beef producti on systems I. Feedlot performance and producti on costs. Livest. Prod. Sci. 61: 213- 225 Food and Agriculture Organizati on of the United Nati ons. (2006) Livestock’s Long Shadow - Environmental Issues and Opti ons. FAO, Rome Food and Agriculture Organizati on of the United Nati ons. (2009a) How to Feed the World in 2050. FAO, Rome, Italy Food and Agriculture Organizati on of the United Nati ons. (2009b) The State of Food Insecurity in the World 2009. FAO, Rome, Italy Food and Agriculture Organizati on of the United Nati ons. (2010) Greenhouse Gas Emissions from the Dairy Sector: A Life Cycle Assessment. FAO, Rome, Italy Keyzer, M.A., Merbis, M.D., Pavel, I.F.P.W., van Wesenbeeck, C.F.A. (2005) Diet shift s towards meat and the eff ects on cereal use: can we feed the animals in 2030? Ecol. Econ. 55: 187-202 U.S. Census Bureau. (2008) Total Midyear Populati on for the World: 1950-2050. htt p://www.census.gov/ipc/www/idb/ worldpop.html Date Accessed: July 2009 USDA. (1978) Livestock Slaughter Annual Summary 1977. USDA, Washington, DC USDA. (2000) Part I: Baseline Reference of Feedlot Management Practi ces, 1999. USDA:APHIS:VS, CEAH, Nati onal Animal Health Monitoring System, Fort Collins, CO USDA. (2005) Dietary Guidelines for Americans 2005. USDA, Washington, DC USDA. (2007) Dairy 2007, Part I: Reference of Dairy Catt le Health and Management Practi ces in the United States, 2007. USDA-APHIS-VS, Fort Collins, CO USDA. (2011) Data and Stati sti cs. htt p://www.nass.usda.gov/ Data_and_Stati sti cs/Quick_Stats/index.asp Date Accessed: May 2011 USDA/NASS. (2008) Livestock Slaughter 2007 Summary. USDA, Washington, DC