Reprod Dom Anim 49 (Suppl. 3), 16–26 (2014); doi: 10.1111/rda.12324 ISSN 0936–6768
Review Article Use of Antimicrobials in the Treatment of Reproductive Diseases in Cattle and Horses S Py€or€al€a, J Taponen and T Katila Department of Production Animal Medicine, Faculty of Veterinary Medicine, University of Helsinki, Saarentaus, Finland
Contents Use of antimicrobials for veterinary indications related to reproduction in cattle and horses is reviewed. Antimicrobial compounds are widely used to treat and prevent infections of reproductive organs. Total amounts of antimicrobials for such purposes, estimated by weight, are low compared with major uses in food animals. The most common reproduction-related indication in cattle is mastitis. The number of intramammary products available for treatment of mastitis in the European Union is high. Metritis and endometritis also require antimicrobial treatment of cattle and specific products for intrauterine administration are available. The traditions and practices associated with the use of these products vary considerably among different countries. Parenteral antimicrobial treatment is used to treat acute clinical mastitis and puerperal metritis. Pharmacological characteristics of the antimicrobial administered parenterally are critical to achieve and maintain therapeutic concentrations in the target organs. In mares, the most common indications associated with reproduction are endometritis, retained placenta and placentitis. The number of authorized antimicrobial products for horses is limited. Horses are treated individually and off-label use of antimicrobials is very common. In veterinary indications related to reproduction, treatment practices exist that cannot be considered to be evidence-based or responsible use of antimicrobials. Not all products for local treatment have proven efficacy data. Examples of unnecessary uses are routine treatment of cows with retained placenta and use of post-breeding antibiotic treatments in mares.
Introduction Antimicrobials are widely used in animals for indications related to reproduction. In cattle, by far the most common indication is treatment of bovine mastitis (Mitchell et al. 1998; Thomson et al. 2008; Menendez Gonz alez et al. 2010). Infections of the uterus, that is, metritis and endometritis, are frequently treated with antimicrobials. Antimicrobials are also routinely used to prevent infections of the reproductive organs. The most common preventive uses are intramammary (IMM) antimicrobial treatment of dairy cows at drying-off and intrauterine (IU) administration of antibiotics in association with retained placenta. A large number of authorized products are available for these indications for cattle in the European Union (E.U.). The availability of authorized products for horses is limited compared with other food animal species. Horses are individually treated and the value of the mare and the foal may justify extensive therapeutic approaches. Horses are considered to be food-producing animals in the E.U., but the value of the meat is trivial
and off-label treatments, and even products intended for humans, are often used (Weese et al. 2006). Antimicrobial therapy related to reproduction, as for all antimicrobial therapy, should only be used when it is unequivocally indicated (OIE 2013). Treatment should preferably be based on pathogen isolation and in vitro susceptibility testing, but this is not always feasible in practical conditions. In vitro susceptibility of the causal agents should be a pre-requisite for treatment, but does not secure treatment response in vivo (Doern and Brecher 2011). If the causal agent is susceptible to the so-termed first-line antimicrobials, they should be used for treatment (Constable et al. 2008; OIE 2013). Fluoroquinolones and 3rd and 4th generation cephalosporins are of special interest from a public health perspective, and their use should be restricted to selected cases only (EMEA 2006; Collignon et al. 2009; Greko et al. 2009). Use of combinations of antimicrobial agents should always be scientifically justified. All treatments should be evidence-based, that is, the efficacy of each product for specific indications should be demonstrated by scientific studies (Cockcroft and Holmes 2003; Hunter et al. 2010). However, efficacy of many old products has not been demonstrated to a similar standard as in case of the more recently authorized drugs. Treatment practices exist that cannot be considered to be evidence based. Use of all antimicrobials selects for resistance, which can emerge among target pathogens and also in commensal microbiota exposed to the antimicrobials (Hunter et al. 2010; Marshall and Levy 2011). Treatment of animals can also directly and indirectly affect public health (OIE 2013). In the treatment of infections of the reproductive organs, antibiotics are mostly administered locally, and doses are smaller than for parenteral use. The exposure of commensal microbiota is thus more limited, and local use can be regarded as less risky compared with parenteral treatment (EMA 2007). In this study, we review the use of antibiotics to treat and prevent commonly occurring infections related to reproduction of cattle and horses, and seek evidencebased, best practice treatment recommendations. Indications, such as treatment of rare infections and preventive uses associated with reproductive surgery or biotechnology, are not covered.
Pharmacological considerations Infections of the reproductive tract can be treated via parenteral, IMM or IU routes. Availability of different © 2014 Blackwell Verlag GmbH
Use of Antimicrobials
antimicrobial boluses and suspensions for IU use varies considerably among E.U. member states. Parenteral treatment is recommended for severe infections with systemic signs and risk of sepsis. The susceptibility of horses to gastrointestinal disturbances limits the selection of compounds that can be used parenterally (Giguere and Afonso 2013). Highly lipophilic antimicrobials, such as macrolides, trimethoprim or fluoroquinolones, have the largest volumes of distribution and produce high concentrations also in peripheral parts of the body (Baggot and Giguere 2013). Compounds with moderate lipid solubility, for instance, sulphonamides and oxytetracycline (OTC), can also yield therapeutic concentrations in the reproductive organs (Parkinson 2009). Penicillins and cephalosporins are mostly ionized in the plasma and their distribution is limited, but against bacterial species with very low minimum inhibitory concentrations they may be used parenterally (Witte et al. 2010; Martinez et al. 2013). A recent study showed that a single subcutaneous administration of ceftiofur crystalline-free acid (CCFA) did not yield therapeutic concentrations in uterine tissues in cows with acute post-partum (PP) metritis (von Krueger et al. 2013). Only very few antimicrobials diffuse to the mammary gland to the extent that therapeutic concentrations are maintained in the milk (Erskine et al. 2003; Constable et al. 2008; Baggot and Giguere 2013). Intracellular penetration of penicillins and aminoglycosides is poor and efficacy against pathogens residing inside the cells (e.g. phagocytes or epithelial cells) is also weak (Martinez et al. 2013). Penetration of antimicrobials into a pregnant uterus has to be considered when treating conditions like placentitis in mares (Santschi and Papich 2000; Weese et al. 2006; Baggot and Giguere 2013). Benzylpenicillin and gentamicin were efficiently transferred to the placenta in pregnant mares (Murchie et al. 2006). Enrofloxacin and trimethoprim–sulphonamides were reported to produce therapeutic concentrations in the endometrium of healthy mares (Brown et al. 1988; Fumuso et al. 2002; Gonz alez et al. 2010). Ceftiofur did not yield therapeutic concentrations in the placenta or foetal fluid of mares with placentitis (Macpherson et al. 2012). Some compounds (fluoroquinolones,
Fig. 1. Distribution of sales of veterinary antimicrobial agents for food-producing animals (including horses) in mg per population correction unit (mg/PCU), by pharmaceutical form in 25 European countries in 2011 (EMA 2013)
© 2014 Blackwell Verlag GmbH
17
tetracyclines) are contraindicated due to toxic effects in the foetus (Baggot and Giguere 2013). In the treatment via local routes, that is, administering the drug directly into the udder or uterus, high concentrations at the infection site are easily achieved (Gilbert and Schwark 1992; Erskine et al. 2003). For IU infusions in the mare, only non-irritating water-soluble antibiotics can be used (Orsini et al. 1996; Parkinson 2009). In the mare, IU treatment is possible only during oestrus, as all manipulations during dioestrus readily result in multiplication of contaminating bacteria. Environmental factors such as the presence of milk or inflammatory secretions at the infection site may interfere with the activity of the drug and must be taken into account (Martinez et al. 2013). Penicillins, aminoglycosides and trimethoprim–sulphonamides may undergo enzymatic degradation or lose activity in the presence of tissue debris and pus (Parkinson 2009; Apley and Coetzee 2013). Aminoglycosides are most active in alkaline environments and are completely inactive under anaerobic conditions (Parkinson 2009; Martinez et al. 2013). Tetracyclines form chelates with casein, which considerably decreases their activity in the milk (Kuang et al. 2009). Sales and use patterns Data on sales of antimicrobial products intended for indications related to reproduction are available for Europe (EMA 2013) (Fig. 1). Total amounts used for local treatment in reproductive indications are marginal compared with other routes: the major part of the sales was sold for oral use and only 7% as injectables and 1% for local uses (IMM and IU preparations). Of the overall sales, the largest group was tetracyclines, which accounted for 37% of the total use, but only 0.3% as IMM and IU preparations. The share of injectables used to treat reproductive infections may be significant but mostly remains unknown; limited data are available on the use in cattle (Thomson et al. 2008; Menendez Gonzalez et al. 2010). For penicillins, IU preparations accounted for 1.5% of the total sales, and only 0.1% was boluses and other
18
uterine preparations (EMA 2013). If the consumption were expressed as daily doses, the proportion of IMM and IU treatments would be greater in all antimicrobial groups where these uses are common. The proportion of IU use is highest for cephalosporins, most of them first or second generation compounds (Fig. 2). Third and fourth generation cephalosporins are an exception in use patterns, as the average proportion of IMM use is as high as 17.2% of the total use. Benzylpenicillin is the most common antibiotic for IMM use in the Nordic countries (NORM-VET 2000–2012; SVARM 2002– 2012); worldwide, the most popular compounds to treat mastitis are aminopenicillins combined with cloxacillin or clavulanic acid, cephalosporins and natural penicillins combined with aminoglycosides (Ruegg 2010; Barlow 2011). In many European countries, blanket dry cow therapy (DCT), where all cows are treated at drying-off, is commonly used. However, there is a general pressure to move from blanket DCT towards selective DCT, to reduce routine use of antibiotics (Cameron et al. 2014; Mevius and Heederik 2014). Beta-lactams in combination with aminoglycosides or cephalosporins are the most common classes of antibiotics used worldwide in DCT (Robert et al. 2006; Barlow 2011).
Antimicrobial use in Cattle Bovine mastitis The most common route of administration for the treatment of mastitis is the IMM route (Ruegg 2010). In the Nordic countries, treatment incidences have been monitored as national health recordings. In 2008, the incidence of treatment of clinical mastitis was the highest (42%) in Denmark and lowest (23%) in Norway (Wolff et al. 2012). In Finland, the annual number of IMM products used for lactating cows in 2012 was two tubes per cow (FIMEA 2013). This corresponds to approximately 35% of the cows being treated during lactation. In an Irish study, the incidence of clinical mastitis was estimated to be on average 54 cases per 100 cow-years at risk (More et al. 2012). In a recent report from the UK, use of IMM antibiotics for lactation therapy was 0.95 grams per dairy cow, but it was difficult to estimate the number of tubes used per cow (VMD 2013).
S Py€ or€ al€ a, J Taponen and T Katila
The most common bacteria causing bovine mastitis are streptococci and staphylococci (Staphylococcus aureus and coagulase-negative staphylococci, CNS) (Barlow 2011). Streptococci have remained fully susceptible to benzylpenicillin (Pitk€al€a et al. 2004; Hendriksen et al. 2008; VMD 2013) but resistance is common among mastitis-causing staphylococci (Pitk€ al€ a et al. 2004; Barlow 2011; VMD 2013). Macrolide resistance has emerged in streptococcal and staphylococcal species of bovine origin (Hendriksen et al. 2008; VMD 2013) and acquired antimicrobial resistance is common among mastitis-causing coliform bacteria. Most common resistance has been reported to OTC and sulphonamides, but it is still rare against fluoroquinolones (Suojala et al. 2013; VMD 2013). In general, antimicrobial resistance is not yet a concern in mastitis treatment, as the main pathogens are susceptible to most compounds available for treatment. Treatment of mastitis should, when possible, be based on bacteriological diagnosis (Constable et al. 2008; Barlow 2011). Sampling for bacteriological diagnosis is recommended in mastitis cases, but in many countries, taking milk samples is not routine. Rapid bacteriological diagnosis would facilitate the correct selection of the antimicrobial. Benzylpenicillin is the drug of choice for the treatment of clinical mastitis caused by penicillinsusceptible bacteria (Constable et al. 2008; Thomson et al. 2008; Barlow 2011). In cases due to penicillinresistant organisms, other beta-lactams are recommended. Treatment of subclinical mastitis is generally not economic during lactation and can be postponed until drying-off (Robert et al. 2006; Barlow 2011). In subclinical mastitis due to very contagious organisms, such as Streptococcus agalactiae, immediate treatment with benzylpenicillin is warranted (Barlow 2011). Intramammary administration is recommended for clinical mastitis caused by streptococcal species (Erskine et al. 2003; Barlow 2011; Kalmus et al. 2014). Streptococci reside in the milk compartment, and there are no pharmacokinetic grounds for parenteral administration, which is more invasive and requires significantly increased dose of the antimicrobial. In mastitis due to S. aureus, parenteral or combination treatment has been suggested to be more efficient (Barlow 2011; Kalmus et al. 2014). If the isolate produces b-lactamase
Fig. 2. Distribution of sales by pharmaceutical form for 1st and 2nd generation cephalosporins in mg/PCU in 22 European countries in 2011 (EMA 2013). No sales were reported in Iceland, Norway and Sweden. In addition, negligible amounts were sold as oral solution in some countries
© 2014 Blackwell Verlag GmbH
Use of Antimicrobials
(penicillin-resistance), response to treatment is poor and drying-off the quarter or culling of the cow is recommended (Constable et al. 2008; Barlow 2011). Scientific evidence for the efficacy of antimicrobial treatment in clinical mastitis caused by Escherichia coli is lacking (Suojala et al. 2013), and therefore supportive treatment using non-steroidal anti-inflammatory drugs (NSAID) is in most cases the treatment of choice. In severe coliform mastitis, use of parenteral antimicrobial treatment to reduce bacterial counts and to rescue the cow is necessary, but IMM treatment of coliform mastitis cannot be recommended. Herd treatment protocols using on-farm diagnostics and delayed treatment for mild to moderate cases of mastitis during lactation has gained popularity in the USA and Canada. In these protocols, antimicrobial treatment is targeted at Grampositive bacteria and a non-antibiotic treatment is used as the first approach to treat mastitis caused by Gramnegative bacteria (Lago et al. 2011). Bovine uterine infections The bovine uterus is almost always contaminated with bacteria during the first days PP (Sheldon et al. 2004; Williams 2013). The most common bacterial species present during the first 2 weeks PP are E. coli, Trueperella pyogenes, Pseudomonas spp., Streptococcus spp., Staphylococcus spp., Clostridium spp., Fusobacterium spp. and Bacteroides spp. (Sheldon et al. 2008; LeBlanc et al. 2011). Establishment of uterine infections depends on the presence of pre-disposing factors, the efficacy of host defence mechanisms and the pathogenicity of the bacteria involved (Silva et al. 2008; Williams 2013). Under field conditions, bacteriological sampling of the uterus is usually not feasible, but selection of the antimicrobial is made on an empirical basis. In studies where bacteriological samples have been taken from the uterus, the clinical significance of the presence of each organism is difficult to assess (Silva et al. 2008, 2009). The new, culture-independent metagenomic methods have uncovered a large number species not reported in bovine uterine samples before, but again, their clinical relevance is completely unknown (Santos et al. 2011). Uterine infection pre-dominated by E. coli during the 1st week PP and T. pyogenes during the 2nd week is associated with subsequent endometritis (LeBlanc 2008), and infection derived from T. pyogenes also with chronic uterine inflammation (Sheldon et al. 2008; Silva et al. 2008). This opportunistic species is frequently present with many other bacterial species but most often with anaerobic species F. necrophorum and Prevotella melaninogenicus, which seem to have a synergistic effect in the pathogenesis of IU infection (Sheldon et al. 2008). All uterine infections can impair reproductive performance, but in particular those caused by T. pyogenes (LeBlanc 2008). Few data on in vitro susceptibilities of bacteria isolated from the bovine uterus are available. E. coli isolated from the uterus are mostly of commensal type and could be compared with isolates from the bovine udder (Silva et al. 2009). In principle, T. pyogenes and most anaerobic species are susceptible to benzylpenicillin and ampicillin in vitro (Jousimies-Somer et al. 1996; VMD 2013). In the UK, nearly half of the © 2014 Blackwell Verlag GmbH
19
T. pyogenes isolates from bovine mastitis were resistant to OTC (VMD 2013), whereas in Finland, no resistance was detected (Jousimies-Somer et al. 1996). Reproductive diseases in cattle caused by bacterial infections usually localize in the tubular genital tract, mainly in the uterus, and thus are commonly termed ‘uterine disease’. Generally, uterine disease represents the only use of antimicrobials for indications related to bovine reproduction and can have detrimental effects on the reproductive performance (Borsberry and Dobson 1989; Huszenicza et al. 1999; LeBlanc et al. 2002a; Gilbert et al. 2005; Williams 2013). Retained placenta (RP), often treated with antimicrobials, is primarily not an infectious disease, but RP is a particularly important pre-disposing factor for uterine infection (Sheldon et al. 2008). Uterine disease is a group of infectious diseases that have different manifestations. Sheldon et al. (2006) recommended a definition of PP uterine disease in cattle, which is now generally used. Bovine puerperal uterine disease is divided into four (or five) different manifestations: puerperal metritis (puerperal metritis and clinical metritis), clinical endometritis, subclinical endometritis and pyometra. In addition to these, RP can be defined as failure to pass the placenta within 24 h of calving (Kelton et al. 1998). From the clinical viewpoint metritis needs to be divided into two different categories of severity: puerperal metritis, with signs of systemic illness, and clinical metritis, without them (also termed puerperal endometritis). Clinical endometritis is defined as the presence of purulent uterine discharge after 20 days or mucopurulent discharge after 26 days PP (Sheldon et al. 2006). Endometrial cytology does not, however, correlate well with the clinical signs (Dubuc et al. 2010a), and these two states seem to have different associated risk factors (Dubuc et al. 2010b). The term ‘purulent vaginal discharge’ could be used instead of clinical endometritis (Dubuc et al. 2010a). A definition of subclinical endometritis was suggested in 2004 (Kasimanickam et al. 2004). At present, diagnostic criteria are still under discussion, and consequently the assessment of treatment effects is difficult. Here, we focus on the antimicrobial treatment of four major types of bovine uterine disease: RP, puerperal metritis, clinical metritis and clinical endometritis. Other diagnoses, such as subclinical endometritis or pyometra, or the use of antimicrobials in reproduction biotechnology, are not discussed. Retained placenta The reported incidences of RP have varied from 1.6% to 15% (Parkinson 2009). Based on a total of fifty reports, the median incidence of RP was 8.6% (Kelton et al. 1998). RP is important because it substantially increases the risk for metritis and endometritis. Retained placenta has been traditionally treated by manual removal, with or without local antimicrobial application. No benefits, or even some harmful effects, of this treatment have been demonstrated (see review by Peters and Laven 1996). Fever is a usual consequence of RP (Stevens et al. 1995; Drillich et al. 2003,
20
S Py€ or€ al€ a, J Taponen and T Katila
2006b), and reduction of body temperature is one objective of the treatments, another being to decrease the risk for metritis and endometritis. In many countries, the routine practice for the treatment of RP has been, and remains, application of antimicrobial boluses in the uterus. The impact of these treatments on subsequent fertility has been inconclusive: some studies showed a benefit, but others no effects or even negative effects (Parkinson 2009). Daily IU infusion of OTC for as long as the placenta was retained reduced the incidence of fever but failed to affect positively the reproductive performance (Stevens et al. 1995). Treatment with chlortetracycline IU twice weekly for 2 weeks had no effects on reproductive performance or milk production (Goshen and Shpigel 2006). In one study, parenteral treatment with ceftiofur for 3(–5) days was compared with local treatment with ampicillin and cloxacillin, in combination with manual removal of the placenta, with no evident difference in reproductive performance (Drillich et al. 2003). Invariably there is minimal evidence of beneficial effects of local antimicrobial treatments of RP on subsequent reproductive performance. Parenteral treatment of cows with RP has also been studied. Ceftiofur hydrochloride for 5 days was compared with no treatment, but no positive effects were reported (Risco and Hernandez 2003). In another study, preventive 3(–5)-day ceftiofur treatment of all cows with RP was compared with the same treatment only provided to cows with fever (Drillich et al. 2006a). No differences in the pregnancy rates were established between the two approaches. A review by LeBlanc (2008) concluded that recent studies consistently indicate that routine parenteral antimicrobial treatment of all cows with RP is not justified and that it is preferable to treat cows that develop metritis.
Antimicrobials recommended for treatment of puerperal metritis include benzylpenicillin, OTC and ceftiofur (Smith et al. 1998; Drillich et al. 2001; Constable et al. 2008). No difference was established between the efficacy of benzylpenicillin and ceftiofur (Smith et al. 1998). Parenteral OTC was effective for treating puerperal metritis (Schmitt et al. 2001).
Puerperal metritis The reported incidence of puerperal metritis varies among countries. It has been suggested that 25–40% of cows have uterine disease in the first 2 weeks after calving (Sheldon et al. 2008), but there is a grey zone between puerperal and clinical metritis causing conflicting results. In some countries, such as Finland, the incidence seems to be much lower; according to national dairy herd statistics, only 1–2% of cows are treated annually for puerperal metritis. Treatment of puerperal metritis is necessary for cow welfare reasons and to reduce the severity of the disease (LeBlanc 2008). Despite the clinical cure, 40–50% of the affected cows suffer later from clinical endometritis (Drillich et al. 2001; Benzaquen et al. 2007). A few decades ago, the treatment of choice was IU infusion of antimicrobials, for example, OTC (Gustafsson 1984; Montes and Pugh 1993) or a combination of ampicillin and cloxacillin (Ahlers et al. 2000), into the uterus. Currently, it is common consensus that cows with puerperal metritis require parenteral antimicrobial treatment (Paisley et al. 1986; LeBlanc 2008). No evidence of any advantage of complementing parenteral with local treatment has been reported (Drillich et al. 2001, 2003).
Clinical endometritis The reported incidence of clinical endometritis (CE) (with various case definitions) has been between 2.2% and 37%, with a median of 10% (Kelton et al. 1998). Sheldon et al. (2008) estimated that uterine disease persists in up to 20% of animals as CE. Clinical endometritis significantly decreases reproductive performance (LeBlanc et al. 2002a). The aim of treatment is to restore reproductive performance of the cow. Based on published literature, Sheldon and Dobson (2004) concluded that administration of PGF is the treatment of choice for CE, at least when a corpus luteum (CL) is present. Nevertheless, in an extensive clinical study, PGF given twice failed to decrease time to pregnancy in cows with CE (Dubuc et al. 2011). Haimerl et al. (2013) concluded in their meta-analysis that PGF treatment did not result in an improvement in reproductive performance in cows with CE. LeBlanc (2008) listed a number of antimicrobial compounds used for IU infusions in cattle, including tetracycline, penicillin, cephapirin, chloramphenicol, diluted Lugol’s iodine, gentamycin, spectinomycin, sulphonamides, nitrofurazone, iodine and chlorhexidine, but many of them are no longer approved for this use. In general, no efficacy data are available to
Clinical metritis The reported incidences of clinical metritis vary, but apparently the majority of the 25–40% of cows with uterine disease in the first 2 weeks after calving suffers from clinical metritis (Sheldon et al. 2008). The self-cure rate for clinical metritis is high (Bekana et al. 1994; Sheldon et al. 2006). The biggest challenge is to detect cows that benefit from antimicrobial treatment (LeBlanc et al. 2002; Sheldon et al. 2006). Studies published on treatment of clinical metritis are scarce. In one Israeli study, cows with clinical metritis (possibly some had puerperal metritis, but no RP cows were included) were treated with chlortetracycline locally twice a week for 2 weeks. The treatment increased milk yield and reduced days open, but the effect was obtained only in pluriparous, not in primiparous cows (Goshen and Shpigel 2006). In cows with RP, immediate treatment with ceftiofur for 5 days decreased the incidence of uterine disease but did not reduce days open (Risco and Hernandez 2003). An alternative to antimicrobial treatment, prostaglandin F2a (PGF) administered during the first three weeks after calving was also ineffective (Kristula and Bartholomew 1998; Hendricks et al. 2006). The conclusion from these studies is that very little, if any, supportive evidence exists for benefits from any treatments of clinical metritis.
© 2014 Blackwell Verlag GmbH
Use of Antimicrobials
support the use of any of them, with the exception of cephapirin. Cows with CE and treated with IU cephapirin between 26 and 33 days PP had a significantly shorter time to pregnancy than untreated cows, but the difference between cephapirin-treated cows and those treated with PGF was not statistically significant (LeBlanc et al. 2002b). In cows with a palpable CL, no significant difference was found between treatments. Two other studies carried out in Australia and New Zealand showed some evidence for improved reproductive performance of cows with purulent vaginal discharge when treated with IU cephapirin approximately 1 month before the first insemination when compared with no treatment (McDougall 2001; Runciman et al. 2009). As an alternative to antimicrobial treatment, the effect of a single IU infusion of dextrose was compared with a single parenteral dose of CCFA or no treatment in cows with CE (Brick et al. 2012). Both treatments increased clinical cure. Cows treated with dextrose tended to conceive better than untreated or CCFA treated CE cows, and they were not significantly different from cows without CE. In another study, parenteral ceftiofur treatment (for 3 days on the 4th week PP) of cows with CE was compared with PGF treatment (two doses two weeks apart) and was found equivalent to PGF (Kaufmann et al. 2010). Supportive evidence of the efficacy of parenteral ceftiofur in the treatment of CE is thus weak. Furthermore, ceftiofur belongs to the group of critically important antibiotics that should be reserved for restricted indications only (Greko et al. 2009). Full consensus on the safest and most efficient and economic treatment of bovine CE is lacking. Antibiotic use in mares Retained placenta and puerperal metritis are serious complications of parturition in horses and require antimicrobial treatment. Pregnant mares are treated with antibiotics for imminent abortions, which are most commonly caused by bacterial placentitis. Endometritis is considered to be the major cause of infertility in mares, but it is probably over diagnosed and treated. Scientific information for efficacy of different antimicrobial treatments of reproductive indications in horses is rare, and treatment recommendations often rely on single studies or empirical evidence. Retained placenta Retained placenta is the most common PP complication in the mare and has been reported to occur between 2% and 10% of births. Antimicrobial treatment is commonly used to prevent metritis, which is a serious complication of RP (Therfall 2011; Giguere and Afonso 2013). If the mare is treated early (≤6 h after parturition) with oxytocin (OT) infusions or injections, manual removal and subsequent high-volume uterine lavage are recommended. This results in full recovery without complications and in high pregnancy rates already after foal heat breedings (Sevinga et al. 2002). © 2014 Blackwell Verlag GmbH
21
If the treatment is delayed, the risk of complications increases. Sequelae of RP are metritis and laminitis, rarely septicaemia and death (Therfall 2011). If the treatment is started late or if OT administrations do not result in the expulsion of the placenta, NSAIDs and antibiotics should be administered. E. coli is the dominant organism in the early PP equine uterus but gradually streptococci begin to overwhelm (Katila et al. 1988). Various antibiotics have been used (tetracyclines, sulphanilamide, benzylpenicillin, polymyxin, amikacin, ticarcillin and others), but no controlled studies have been performed. If the mare is hospitalized, intravenous gentamicin is the drug of choice, in combination with benzylpenicillin. If the mare is in the home stable, per oral trimethoprim–sulphonamide is a practical alternative. Post-partum metritis Metritis
Review Article Use of Antimicrobials in the Treatment of Reproductive Diseases in Cattle and Horses S Py€or€al€a, J Taponen and T Katila Department of Production Animal Medicine, Faculty of Veterinary Medicine, University of Helsinki, Saarentaus, Finland
Contents Use of antimicrobials for veterinary indications related to reproduction in cattle and horses is reviewed. Antimicrobial compounds are widely used to treat and prevent infections of reproductive organs. Total amounts of antimicrobials for such purposes, estimated by weight, are low compared with major uses in food animals. The most common reproduction-related indication in cattle is mastitis. The number of intramammary products available for treatment of mastitis in the European Union is high. Metritis and endometritis also require antimicrobial treatment of cattle and specific products for intrauterine administration are available. The traditions and practices associated with the use of these products vary considerably among different countries. Parenteral antimicrobial treatment is used to treat acute clinical mastitis and puerperal metritis. Pharmacological characteristics of the antimicrobial administered parenterally are critical to achieve and maintain therapeutic concentrations in the target organs. In mares, the most common indications associated with reproduction are endometritis, retained placenta and placentitis. The number of authorized antimicrobial products for horses is limited. Horses are treated individually and off-label use of antimicrobials is very common. In veterinary indications related to reproduction, treatment practices exist that cannot be considered to be evidence-based or responsible use of antimicrobials. Not all products for local treatment have proven efficacy data. Examples of unnecessary uses are routine treatment of cows with retained placenta and use of post-breeding antibiotic treatments in mares.
Introduction Antimicrobials are widely used in animals for indications related to reproduction. In cattle, by far the most common indication is treatment of bovine mastitis (Mitchell et al. 1998; Thomson et al. 2008; Menendez Gonz alez et al. 2010). Infections of the uterus, that is, metritis and endometritis, are frequently treated with antimicrobials. Antimicrobials are also routinely used to prevent infections of the reproductive organs. The most common preventive uses are intramammary (IMM) antimicrobial treatment of dairy cows at drying-off and intrauterine (IU) administration of antibiotics in association with retained placenta. A large number of authorized products are available for these indications for cattle in the European Union (E.U.). The availability of authorized products for horses is limited compared with other food animal species. Horses are individually treated and the value of the mare and the foal may justify extensive therapeutic approaches. Horses are considered to be food-producing animals in the E.U., but the value of the meat is trivial
and off-label treatments, and even products intended for humans, are often used (Weese et al. 2006). Antimicrobial therapy related to reproduction, as for all antimicrobial therapy, should only be used when it is unequivocally indicated (OIE 2013). Treatment should preferably be based on pathogen isolation and in vitro susceptibility testing, but this is not always feasible in practical conditions. In vitro susceptibility of the causal agents should be a pre-requisite for treatment, but does not secure treatment response in vivo (Doern and Brecher 2011). If the causal agent is susceptible to the so-termed first-line antimicrobials, they should be used for treatment (Constable et al. 2008; OIE 2013). Fluoroquinolones and 3rd and 4th generation cephalosporins are of special interest from a public health perspective, and their use should be restricted to selected cases only (EMEA 2006; Collignon et al. 2009; Greko et al. 2009). Use of combinations of antimicrobial agents should always be scientifically justified. All treatments should be evidence-based, that is, the efficacy of each product for specific indications should be demonstrated by scientific studies (Cockcroft and Holmes 2003; Hunter et al. 2010). However, efficacy of many old products has not been demonstrated to a similar standard as in case of the more recently authorized drugs. Treatment practices exist that cannot be considered to be evidence based. Use of all antimicrobials selects for resistance, which can emerge among target pathogens and also in commensal microbiota exposed to the antimicrobials (Hunter et al. 2010; Marshall and Levy 2011). Treatment of animals can also directly and indirectly affect public health (OIE 2013). In the treatment of infections of the reproductive organs, antibiotics are mostly administered locally, and doses are smaller than for parenteral use. The exposure of commensal microbiota is thus more limited, and local use can be regarded as less risky compared with parenteral treatment (EMA 2007). In this study, we review the use of antibiotics to treat and prevent commonly occurring infections related to reproduction of cattle and horses, and seek evidencebased, best practice treatment recommendations. Indications, such as treatment of rare infections and preventive uses associated with reproductive surgery or biotechnology, are not covered.
Pharmacological considerations Infections of the reproductive tract can be treated via parenteral, IMM or IU routes. Availability of different © 2014 Blackwell Verlag GmbH
Use of Antimicrobials
antimicrobial boluses and suspensions for IU use varies considerably among E.U. member states. Parenteral treatment is recommended for severe infections with systemic signs and risk of sepsis. The susceptibility of horses to gastrointestinal disturbances limits the selection of compounds that can be used parenterally (Giguere and Afonso 2013). Highly lipophilic antimicrobials, such as macrolides, trimethoprim or fluoroquinolones, have the largest volumes of distribution and produce high concentrations also in peripheral parts of the body (Baggot and Giguere 2013). Compounds with moderate lipid solubility, for instance, sulphonamides and oxytetracycline (OTC), can also yield therapeutic concentrations in the reproductive organs (Parkinson 2009). Penicillins and cephalosporins are mostly ionized in the plasma and their distribution is limited, but against bacterial species with very low minimum inhibitory concentrations they may be used parenterally (Witte et al. 2010; Martinez et al. 2013). A recent study showed that a single subcutaneous administration of ceftiofur crystalline-free acid (CCFA) did not yield therapeutic concentrations in uterine tissues in cows with acute post-partum (PP) metritis (von Krueger et al. 2013). Only very few antimicrobials diffuse to the mammary gland to the extent that therapeutic concentrations are maintained in the milk (Erskine et al. 2003; Constable et al. 2008; Baggot and Giguere 2013). Intracellular penetration of penicillins and aminoglycosides is poor and efficacy against pathogens residing inside the cells (e.g. phagocytes or epithelial cells) is also weak (Martinez et al. 2013). Penetration of antimicrobials into a pregnant uterus has to be considered when treating conditions like placentitis in mares (Santschi and Papich 2000; Weese et al. 2006; Baggot and Giguere 2013). Benzylpenicillin and gentamicin were efficiently transferred to the placenta in pregnant mares (Murchie et al. 2006). Enrofloxacin and trimethoprim–sulphonamides were reported to produce therapeutic concentrations in the endometrium of healthy mares (Brown et al. 1988; Fumuso et al. 2002; Gonz alez et al. 2010). Ceftiofur did not yield therapeutic concentrations in the placenta or foetal fluid of mares with placentitis (Macpherson et al. 2012). Some compounds (fluoroquinolones,
Fig. 1. Distribution of sales of veterinary antimicrobial agents for food-producing animals (including horses) in mg per population correction unit (mg/PCU), by pharmaceutical form in 25 European countries in 2011 (EMA 2013)
© 2014 Blackwell Verlag GmbH
17
tetracyclines) are contraindicated due to toxic effects in the foetus (Baggot and Giguere 2013). In the treatment via local routes, that is, administering the drug directly into the udder or uterus, high concentrations at the infection site are easily achieved (Gilbert and Schwark 1992; Erskine et al. 2003). For IU infusions in the mare, only non-irritating water-soluble antibiotics can be used (Orsini et al. 1996; Parkinson 2009). In the mare, IU treatment is possible only during oestrus, as all manipulations during dioestrus readily result in multiplication of contaminating bacteria. Environmental factors such as the presence of milk or inflammatory secretions at the infection site may interfere with the activity of the drug and must be taken into account (Martinez et al. 2013). Penicillins, aminoglycosides and trimethoprim–sulphonamides may undergo enzymatic degradation or lose activity in the presence of tissue debris and pus (Parkinson 2009; Apley and Coetzee 2013). Aminoglycosides are most active in alkaline environments and are completely inactive under anaerobic conditions (Parkinson 2009; Martinez et al. 2013). Tetracyclines form chelates with casein, which considerably decreases their activity in the milk (Kuang et al. 2009). Sales and use patterns Data on sales of antimicrobial products intended for indications related to reproduction are available for Europe (EMA 2013) (Fig. 1). Total amounts used for local treatment in reproductive indications are marginal compared with other routes: the major part of the sales was sold for oral use and only 7% as injectables and 1% for local uses (IMM and IU preparations). Of the overall sales, the largest group was tetracyclines, which accounted for 37% of the total use, but only 0.3% as IMM and IU preparations. The share of injectables used to treat reproductive infections may be significant but mostly remains unknown; limited data are available on the use in cattle (Thomson et al. 2008; Menendez Gonzalez et al. 2010). For penicillins, IU preparations accounted for 1.5% of the total sales, and only 0.1% was boluses and other
18
uterine preparations (EMA 2013). If the consumption were expressed as daily doses, the proportion of IMM and IU treatments would be greater in all antimicrobial groups where these uses are common. The proportion of IU use is highest for cephalosporins, most of them first or second generation compounds (Fig. 2). Third and fourth generation cephalosporins are an exception in use patterns, as the average proportion of IMM use is as high as 17.2% of the total use. Benzylpenicillin is the most common antibiotic for IMM use in the Nordic countries (NORM-VET 2000–2012; SVARM 2002– 2012); worldwide, the most popular compounds to treat mastitis are aminopenicillins combined with cloxacillin or clavulanic acid, cephalosporins and natural penicillins combined with aminoglycosides (Ruegg 2010; Barlow 2011). In many European countries, blanket dry cow therapy (DCT), where all cows are treated at drying-off, is commonly used. However, there is a general pressure to move from blanket DCT towards selective DCT, to reduce routine use of antibiotics (Cameron et al. 2014; Mevius and Heederik 2014). Beta-lactams in combination with aminoglycosides or cephalosporins are the most common classes of antibiotics used worldwide in DCT (Robert et al. 2006; Barlow 2011).
Antimicrobial use in Cattle Bovine mastitis The most common route of administration for the treatment of mastitis is the IMM route (Ruegg 2010). In the Nordic countries, treatment incidences have been monitored as national health recordings. In 2008, the incidence of treatment of clinical mastitis was the highest (42%) in Denmark and lowest (23%) in Norway (Wolff et al. 2012). In Finland, the annual number of IMM products used for lactating cows in 2012 was two tubes per cow (FIMEA 2013). This corresponds to approximately 35% of the cows being treated during lactation. In an Irish study, the incidence of clinical mastitis was estimated to be on average 54 cases per 100 cow-years at risk (More et al. 2012). In a recent report from the UK, use of IMM antibiotics for lactation therapy was 0.95 grams per dairy cow, but it was difficult to estimate the number of tubes used per cow (VMD 2013).
S Py€ or€ al€ a, J Taponen and T Katila
The most common bacteria causing bovine mastitis are streptococci and staphylococci (Staphylococcus aureus and coagulase-negative staphylococci, CNS) (Barlow 2011). Streptococci have remained fully susceptible to benzylpenicillin (Pitk€al€a et al. 2004; Hendriksen et al. 2008; VMD 2013) but resistance is common among mastitis-causing staphylococci (Pitk€ al€ a et al. 2004; Barlow 2011; VMD 2013). Macrolide resistance has emerged in streptococcal and staphylococcal species of bovine origin (Hendriksen et al. 2008; VMD 2013) and acquired antimicrobial resistance is common among mastitis-causing coliform bacteria. Most common resistance has been reported to OTC and sulphonamides, but it is still rare against fluoroquinolones (Suojala et al. 2013; VMD 2013). In general, antimicrobial resistance is not yet a concern in mastitis treatment, as the main pathogens are susceptible to most compounds available for treatment. Treatment of mastitis should, when possible, be based on bacteriological diagnosis (Constable et al. 2008; Barlow 2011). Sampling for bacteriological diagnosis is recommended in mastitis cases, but in many countries, taking milk samples is not routine. Rapid bacteriological diagnosis would facilitate the correct selection of the antimicrobial. Benzylpenicillin is the drug of choice for the treatment of clinical mastitis caused by penicillinsusceptible bacteria (Constable et al. 2008; Thomson et al. 2008; Barlow 2011). In cases due to penicillinresistant organisms, other beta-lactams are recommended. Treatment of subclinical mastitis is generally not economic during lactation and can be postponed until drying-off (Robert et al. 2006; Barlow 2011). In subclinical mastitis due to very contagious organisms, such as Streptococcus agalactiae, immediate treatment with benzylpenicillin is warranted (Barlow 2011). Intramammary administration is recommended for clinical mastitis caused by streptococcal species (Erskine et al. 2003; Barlow 2011; Kalmus et al. 2014). Streptococci reside in the milk compartment, and there are no pharmacokinetic grounds for parenteral administration, which is more invasive and requires significantly increased dose of the antimicrobial. In mastitis due to S. aureus, parenteral or combination treatment has been suggested to be more efficient (Barlow 2011; Kalmus et al. 2014). If the isolate produces b-lactamase
Fig. 2. Distribution of sales by pharmaceutical form for 1st and 2nd generation cephalosporins in mg/PCU in 22 European countries in 2011 (EMA 2013). No sales were reported in Iceland, Norway and Sweden. In addition, negligible amounts were sold as oral solution in some countries
© 2014 Blackwell Verlag GmbH
Use of Antimicrobials
(penicillin-resistance), response to treatment is poor and drying-off the quarter or culling of the cow is recommended (Constable et al. 2008; Barlow 2011). Scientific evidence for the efficacy of antimicrobial treatment in clinical mastitis caused by Escherichia coli is lacking (Suojala et al. 2013), and therefore supportive treatment using non-steroidal anti-inflammatory drugs (NSAID) is in most cases the treatment of choice. In severe coliform mastitis, use of parenteral antimicrobial treatment to reduce bacterial counts and to rescue the cow is necessary, but IMM treatment of coliform mastitis cannot be recommended. Herd treatment protocols using on-farm diagnostics and delayed treatment for mild to moderate cases of mastitis during lactation has gained popularity in the USA and Canada. In these protocols, antimicrobial treatment is targeted at Grampositive bacteria and a non-antibiotic treatment is used as the first approach to treat mastitis caused by Gramnegative bacteria (Lago et al. 2011). Bovine uterine infections The bovine uterus is almost always contaminated with bacteria during the first days PP (Sheldon et al. 2004; Williams 2013). The most common bacterial species present during the first 2 weeks PP are E. coli, Trueperella pyogenes, Pseudomonas spp., Streptococcus spp., Staphylococcus spp., Clostridium spp., Fusobacterium spp. and Bacteroides spp. (Sheldon et al. 2008; LeBlanc et al. 2011). Establishment of uterine infections depends on the presence of pre-disposing factors, the efficacy of host defence mechanisms and the pathogenicity of the bacteria involved (Silva et al. 2008; Williams 2013). Under field conditions, bacteriological sampling of the uterus is usually not feasible, but selection of the antimicrobial is made on an empirical basis. In studies where bacteriological samples have been taken from the uterus, the clinical significance of the presence of each organism is difficult to assess (Silva et al. 2008, 2009). The new, culture-independent metagenomic methods have uncovered a large number species not reported in bovine uterine samples before, but again, their clinical relevance is completely unknown (Santos et al. 2011). Uterine infection pre-dominated by E. coli during the 1st week PP and T. pyogenes during the 2nd week is associated with subsequent endometritis (LeBlanc 2008), and infection derived from T. pyogenes also with chronic uterine inflammation (Sheldon et al. 2008; Silva et al. 2008). This opportunistic species is frequently present with many other bacterial species but most often with anaerobic species F. necrophorum and Prevotella melaninogenicus, which seem to have a synergistic effect in the pathogenesis of IU infection (Sheldon et al. 2008). All uterine infections can impair reproductive performance, but in particular those caused by T. pyogenes (LeBlanc 2008). Few data on in vitro susceptibilities of bacteria isolated from the bovine uterus are available. E. coli isolated from the uterus are mostly of commensal type and could be compared with isolates from the bovine udder (Silva et al. 2009). In principle, T. pyogenes and most anaerobic species are susceptible to benzylpenicillin and ampicillin in vitro (Jousimies-Somer et al. 1996; VMD 2013). In the UK, nearly half of the © 2014 Blackwell Verlag GmbH
19
T. pyogenes isolates from bovine mastitis were resistant to OTC (VMD 2013), whereas in Finland, no resistance was detected (Jousimies-Somer et al. 1996). Reproductive diseases in cattle caused by bacterial infections usually localize in the tubular genital tract, mainly in the uterus, and thus are commonly termed ‘uterine disease’. Generally, uterine disease represents the only use of antimicrobials for indications related to bovine reproduction and can have detrimental effects on the reproductive performance (Borsberry and Dobson 1989; Huszenicza et al. 1999; LeBlanc et al. 2002a; Gilbert et al. 2005; Williams 2013). Retained placenta (RP), often treated with antimicrobials, is primarily not an infectious disease, but RP is a particularly important pre-disposing factor for uterine infection (Sheldon et al. 2008). Uterine disease is a group of infectious diseases that have different manifestations. Sheldon et al. (2006) recommended a definition of PP uterine disease in cattle, which is now generally used. Bovine puerperal uterine disease is divided into four (or five) different manifestations: puerperal metritis (puerperal metritis and clinical metritis), clinical endometritis, subclinical endometritis and pyometra. In addition to these, RP can be defined as failure to pass the placenta within 24 h of calving (Kelton et al. 1998). From the clinical viewpoint metritis needs to be divided into two different categories of severity: puerperal metritis, with signs of systemic illness, and clinical metritis, without them (also termed puerperal endometritis). Clinical endometritis is defined as the presence of purulent uterine discharge after 20 days or mucopurulent discharge after 26 days PP (Sheldon et al. 2006). Endometrial cytology does not, however, correlate well with the clinical signs (Dubuc et al. 2010a), and these two states seem to have different associated risk factors (Dubuc et al. 2010b). The term ‘purulent vaginal discharge’ could be used instead of clinical endometritis (Dubuc et al. 2010a). A definition of subclinical endometritis was suggested in 2004 (Kasimanickam et al. 2004). At present, diagnostic criteria are still under discussion, and consequently the assessment of treatment effects is difficult. Here, we focus on the antimicrobial treatment of four major types of bovine uterine disease: RP, puerperal metritis, clinical metritis and clinical endometritis. Other diagnoses, such as subclinical endometritis or pyometra, or the use of antimicrobials in reproduction biotechnology, are not discussed. Retained placenta The reported incidences of RP have varied from 1.6% to 15% (Parkinson 2009). Based on a total of fifty reports, the median incidence of RP was 8.6% (Kelton et al. 1998). RP is important because it substantially increases the risk for metritis and endometritis. Retained placenta has been traditionally treated by manual removal, with or without local antimicrobial application. No benefits, or even some harmful effects, of this treatment have been demonstrated (see review by Peters and Laven 1996). Fever is a usual consequence of RP (Stevens et al. 1995; Drillich et al. 2003,
20
S Py€ or€ al€ a, J Taponen and T Katila
2006b), and reduction of body temperature is one objective of the treatments, another being to decrease the risk for metritis and endometritis. In many countries, the routine practice for the treatment of RP has been, and remains, application of antimicrobial boluses in the uterus. The impact of these treatments on subsequent fertility has been inconclusive: some studies showed a benefit, but others no effects or even negative effects (Parkinson 2009). Daily IU infusion of OTC for as long as the placenta was retained reduced the incidence of fever but failed to affect positively the reproductive performance (Stevens et al. 1995). Treatment with chlortetracycline IU twice weekly for 2 weeks had no effects on reproductive performance or milk production (Goshen and Shpigel 2006). In one study, parenteral treatment with ceftiofur for 3(–5) days was compared with local treatment with ampicillin and cloxacillin, in combination with manual removal of the placenta, with no evident difference in reproductive performance (Drillich et al. 2003). Invariably there is minimal evidence of beneficial effects of local antimicrobial treatments of RP on subsequent reproductive performance. Parenteral treatment of cows with RP has also been studied. Ceftiofur hydrochloride for 5 days was compared with no treatment, but no positive effects were reported (Risco and Hernandez 2003). In another study, preventive 3(–5)-day ceftiofur treatment of all cows with RP was compared with the same treatment only provided to cows with fever (Drillich et al. 2006a). No differences in the pregnancy rates were established between the two approaches. A review by LeBlanc (2008) concluded that recent studies consistently indicate that routine parenteral antimicrobial treatment of all cows with RP is not justified and that it is preferable to treat cows that develop metritis.
Antimicrobials recommended for treatment of puerperal metritis include benzylpenicillin, OTC and ceftiofur (Smith et al. 1998; Drillich et al. 2001; Constable et al. 2008). No difference was established between the efficacy of benzylpenicillin and ceftiofur (Smith et al. 1998). Parenteral OTC was effective for treating puerperal metritis (Schmitt et al. 2001).
Puerperal metritis The reported incidence of puerperal metritis varies among countries. It has been suggested that 25–40% of cows have uterine disease in the first 2 weeks after calving (Sheldon et al. 2008), but there is a grey zone between puerperal and clinical metritis causing conflicting results. In some countries, such as Finland, the incidence seems to be much lower; according to national dairy herd statistics, only 1–2% of cows are treated annually for puerperal metritis. Treatment of puerperal metritis is necessary for cow welfare reasons and to reduce the severity of the disease (LeBlanc 2008). Despite the clinical cure, 40–50% of the affected cows suffer later from clinical endometritis (Drillich et al. 2001; Benzaquen et al. 2007). A few decades ago, the treatment of choice was IU infusion of antimicrobials, for example, OTC (Gustafsson 1984; Montes and Pugh 1993) or a combination of ampicillin and cloxacillin (Ahlers et al. 2000), into the uterus. Currently, it is common consensus that cows with puerperal metritis require parenteral antimicrobial treatment (Paisley et al. 1986; LeBlanc 2008). No evidence of any advantage of complementing parenteral with local treatment has been reported (Drillich et al. 2001, 2003).
Clinical endometritis The reported incidence of clinical endometritis (CE) (with various case definitions) has been between 2.2% and 37%, with a median of 10% (Kelton et al. 1998). Sheldon et al. (2008) estimated that uterine disease persists in up to 20% of animals as CE. Clinical endometritis significantly decreases reproductive performance (LeBlanc et al. 2002a). The aim of treatment is to restore reproductive performance of the cow. Based on published literature, Sheldon and Dobson (2004) concluded that administration of PGF is the treatment of choice for CE, at least when a corpus luteum (CL) is present. Nevertheless, in an extensive clinical study, PGF given twice failed to decrease time to pregnancy in cows with CE (Dubuc et al. 2011). Haimerl et al. (2013) concluded in their meta-analysis that PGF treatment did not result in an improvement in reproductive performance in cows with CE. LeBlanc (2008) listed a number of antimicrobial compounds used for IU infusions in cattle, including tetracycline, penicillin, cephapirin, chloramphenicol, diluted Lugol’s iodine, gentamycin, spectinomycin, sulphonamides, nitrofurazone, iodine and chlorhexidine, but many of them are no longer approved for this use. In general, no efficacy data are available to
Clinical metritis The reported incidences of clinical metritis vary, but apparently the majority of the 25–40% of cows with uterine disease in the first 2 weeks after calving suffers from clinical metritis (Sheldon et al. 2008). The self-cure rate for clinical metritis is high (Bekana et al. 1994; Sheldon et al. 2006). The biggest challenge is to detect cows that benefit from antimicrobial treatment (LeBlanc et al. 2002; Sheldon et al. 2006). Studies published on treatment of clinical metritis are scarce. In one Israeli study, cows with clinical metritis (possibly some had puerperal metritis, but no RP cows were included) were treated with chlortetracycline locally twice a week for 2 weeks. The treatment increased milk yield and reduced days open, but the effect was obtained only in pluriparous, not in primiparous cows (Goshen and Shpigel 2006). In cows with RP, immediate treatment with ceftiofur for 5 days decreased the incidence of uterine disease but did not reduce days open (Risco and Hernandez 2003). An alternative to antimicrobial treatment, prostaglandin F2a (PGF) administered during the first three weeks after calving was also ineffective (Kristula and Bartholomew 1998; Hendricks et al. 2006). The conclusion from these studies is that very little, if any, supportive evidence exists for benefits from any treatments of clinical metritis.
© 2014 Blackwell Verlag GmbH
Use of Antimicrobials
support the use of any of them, with the exception of cephapirin. Cows with CE and treated with IU cephapirin between 26 and 33 days PP had a significantly shorter time to pregnancy than untreated cows, but the difference between cephapirin-treated cows and those treated with PGF was not statistically significant (LeBlanc et al. 2002b). In cows with a palpable CL, no significant difference was found between treatments. Two other studies carried out in Australia and New Zealand showed some evidence for improved reproductive performance of cows with purulent vaginal discharge when treated with IU cephapirin approximately 1 month before the first insemination when compared with no treatment (McDougall 2001; Runciman et al. 2009). As an alternative to antimicrobial treatment, the effect of a single IU infusion of dextrose was compared with a single parenteral dose of CCFA or no treatment in cows with CE (Brick et al. 2012). Both treatments increased clinical cure. Cows treated with dextrose tended to conceive better than untreated or CCFA treated CE cows, and they were not significantly different from cows without CE. In another study, parenteral ceftiofur treatment (for 3 days on the 4th week PP) of cows with CE was compared with PGF treatment (two doses two weeks apart) and was found equivalent to PGF (Kaufmann et al. 2010). Supportive evidence of the efficacy of parenteral ceftiofur in the treatment of CE is thus weak. Furthermore, ceftiofur belongs to the group of critically important antibiotics that should be reserved for restricted indications only (Greko et al. 2009). Full consensus on the safest and most efficient and economic treatment of bovine CE is lacking. Antibiotic use in mares Retained placenta and puerperal metritis are serious complications of parturition in horses and require antimicrobial treatment. Pregnant mares are treated with antibiotics for imminent abortions, which are most commonly caused by bacterial placentitis. Endometritis is considered to be the major cause of infertility in mares, but it is probably over diagnosed and treated. Scientific information for efficacy of different antimicrobial treatments of reproductive indications in horses is rare, and treatment recommendations often rely on single studies or empirical evidence. Retained placenta Retained placenta is the most common PP complication in the mare and has been reported to occur between 2% and 10% of births. Antimicrobial treatment is commonly used to prevent metritis, which is a serious complication of RP (Therfall 2011; Giguere and Afonso 2013). If the mare is treated early (≤6 h after parturition) with oxytocin (OT) infusions or injections, manual removal and subsequent high-volume uterine lavage are recommended. This results in full recovery without complications and in high pregnancy rates already after foal heat breedings (Sevinga et al. 2002). © 2014 Blackwell Verlag GmbH
21
If the treatment is delayed, the risk of complications increases. Sequelae of RP are metritis and laminitis, rarely septicaemia and death (Therfall 2011). If the treatment is started late or if OT administrations do not result in the expulsion of the placenta, NSAIDs and antibiotics should be administered. E. coli is the dominant organism in the early PP equine uterus but gradually streptococci begin to overwhelm (Katila et al. 1988). Various antibiotics have been used (tetracyclines, sulphanilamide, benzylpenicillin, polymyxin, amikacin, ticarcillin and others), but no controlled studies have been performed. If the mare is hospitalized, intravenous gentamicin is the drug of choice, in combination with benzylpenicillin. If the mare is in the home stable, per oral trimethoprim–sulphonamide is a practical alternative. Post-partum metritis Metritis
