In mares with placentitis does the duration of antibiotic treatment affect foal outcome?

PICO question 
In mares with placentitis does treatment with long-term antibiotics result in improved foal viability when compared to repeated short courses of 7 to 10 days? 
  
Clinical bottom line 
Category of research question 
Treatment 
The number and type of study designs reviewed 
The literature search identified six publications that included length of antibiotic treatment and foetal outcome. The publications consisted of four non-randomised non-blinded controlled trials and two randomised non-blinded controlled trials 
Strength of evidence 
Collectively there was weak evidence to support either an intermittent or continuous antibiotic protocol in the treatment of placentitis in mares 
Outcomes reported 
The literature involved experimental induction of ascending placentitis with foal survival or viability as the outcome 
Conclusion 
Further research is required into the diagnosis of placentitis, length of treatment and choice of antibiotic/s to penetrate the uterus in a diseased state 
  
How to apply this evidence in practice 
The application of evidence into practice should take into account multiple factors, not limited to: individual clinical expertise, patient’s circumstances and owners’ values, country, location or clinic where you work, the individual case in front of you, the availability of therapies and resources. 
Knowledge Summaries are a resource to help reinforce or inform decision-making. They do not override the responsibility or judgement of the practitioner to do what is best for the animal in their care. 
  



The evidence
The literature search identified six publications that included antibiotic length of treatment and foetal outcome. The publications consisted of four non-randomised non-blinded controlled trials of level 4 (lower quality controlled trials) (OCEBM Levels of Evidence Working Group) (Macpherson et al., 2013;Christensen et al., 2010;Murchie et al., 2006;and Ryan et al., 2008). This included one pilot study (Ryan et al., 2008) and one short report presented at a conference (Christensen et al., 2010). Two publications were randomised non-blinded controlled trials of level 2b (lower quality randomised controlled trials) (Bailey et al., 2010;and Curcio et al., 2017). All studies involved experimental induction of ascending placentitis with S. zooepidemicus. Pony mares were used in 4/6 studies with one study involving light bred horses (Christensen et al., 2010) and one study using Criollo type mares (Curcio et al., 2017). All horses were in late gestation either unspecified gestational days (Christensen et al., 2010) or between 269 to 300 days gestation (Macpherson et  The outcome consistently measured between studies was 'foal viability', however the parameters determining viability varied between studies. Foal viability was described by Macpherson et al. (2013) and Bailey et al. (2010) as a foal able to right themselves, breathe without assistance, respond to stimuli and suckle. Low risk foals were determined by Curcio et al. (2017) as those able to breathe without assistance in less than 2 minutes, attain sternal recumbency in less than 5 minutes, a normal suckle reflex in 20 minutes, and able to stand with little or no assistance. Foal viability parameters were not defined in two studies (Christensen et al., 2010;and Murchie et al., 2008) and live foal rate was used in a single study (Ryan et al., 2008).
In studies with positive control (infected and not treated) any treatment improved foal outcome compared to no treatment at all (Bailey et al., 2010;and Curcio et al., 2017). The exception is Macpherson et al. (2013) that showed no difference in foal viability in untreated mares and mares receiving CCFA. There was no evidence to support that continuous antibiotics improved foal viability with similar foal viabilities seen in mares treated with TMS for 10 days in Curcio et al. (2017) at 93% (26 viable foals from 28 infected mares that received antibiotics) compared to antibiotics prescribed until abortion or parturition in Ryan     Penicillin G and gentamicin were detected in allantoic fluid Study 2:  Placental drug transfer may be altered if active placental infection is present  Potential for increased dose intervals for penicillin G and increased dose rate of gentamicin to effectively combat placental infections in mares  Both mares that were infected aborted non-viable foals one at 291 days and one at 307 days gestation Limitations:  Small sample size of two mares in Study 2 make conclusions regarding efficacy of 7 days treatment difficult to surmise  Intervention procedure of allantoic sampling may also have compromised foetal well-being with only two out of five viable foals born at 32 and 45 days after experimentation, whilst one mare that had allantocentesis performed and was not inoculated aborted at 315 days gestation 10 mg/mare intramuscular every 3 days for three treatments  Group 2D (n = 6): As Group A with altrenogest as described in Group B and ECP as described in Group C  Group 2E (n = 10): Infected no treatment All treatments were initiated 48 hours after inoculation and continued for 10 days  Time from inoculation to delivery was not significantly different between Group 1 control mares and Group 2D mares receiving supplemental EC (mean 35 and 46 days respectively)  Time from inoculation to delivery was significantly shorter in mares receiving other treatments in Groups 2B, 2C, and 2D compared to control mares. However, delivery was shortest in mares not receiving any treatment (Group 2)  Foal survival at parturition and 7 days of age were similar amongst treated Groups 2A-2E to control (Group 1) ranging between 66.7-100%  There was no significant difference in the number of high risk foals in Group 2D compared to control mares (Group 1)  There was a significant difference in the number of high risk foals in Groups 2A-2C and 2E compared to control (Group 1)  Mares in Group 2E had significantly higher number of dystocia's and premature parturitions

Limitations:
 Unknown if the same mares were used over multiple seasons, which order of treatments were received, and if those inoculated were known to no longer be infected  Unknown if strain cultured was sensitive to TMS in vitro  Small treatment groups

Appraisal, application and reflection
Placentitis is a common condition estimated to affect 3-5% of Thoroughbred pregnancies and can be challenging for the clinician to diagnose and treat (Canisso et al., 2015). Current treatment protocols recommend a multifactorial approach involving antibiotics, anti-inflammatories and progesterone (Murchie et  inoculation to parturition of treated mares 23.7 days and foal survival of 93% (26/28). Bailey et al. (2010) treated until abortion or delivery with average time from inoculation to parturition 31 days with 83% (10/12) of treated mares producing viable foals. Treatment commenced later in the study by Curcio et al. (2017) with the average gestation of mares approximately 300 days, whilst mares in the study by Bailey et al. (2010) were 280-295 days gestation. Viability of the foals and mean gestational lengths were similar between the studies, regardless of duration of antibiotics (both over 320 days gestation). Clinically it is recognised that chronic insidious placentitis carriers a better prognosis compared to acute disease as the foal has time to mature. The aim of placentitis treatment may be to delay parturition and allow foetal maturation as opposed to eliminating bacteria. The study by Ryan et al. (2008) additionally evaluated the use of antibiotics in mares with placentitis after abortion or foaling. Post mortem investigation of the non-viable foals found evidence S. zooepidemicus in foal lungs from mares that were treated with antibiotics and those that were not. The majority of foals in the study by Ryan et al. (2008) born from treated mares had negative blood cultures (10/12 foals treated with antibiotics had negative blood cultures). Uterine cultures taken from mares immediately post foaling or aborting that were on antibiotics were less predictable, with two thirds of mares (8/12 mares) returning a positive uterine culture to S. zooepidemicus despite the administration of antibiotics for over 4 weeks. It was proposed by Ryan et al. (2008) that early initiation of treatment was able to inhibit bacterial growth and subsequent inflammation. As treatment was initiated at onset of clinical signs, delayed antibiotic administration may only suppress bacterial growth. Further research is required into when to start antibiotic treatment and improve detection of mares with suspected placentitis.
Methods to monitor foetal well-being to aid in determining when to treat placentitis is limited. Curcio  correlation of foetal circulation including heart rate, cord pressure (Vincze et al., 2019) and thickness of the placenta measured both abdominally and rectally may aid in earlier identification of mares with placentitis (Curcio et al., 2017). Decreasing progesterone and rising oestrogen levels have been correlated with poor prognosis of foetal survival (Curcio et al., 2017). Allantocentesis samples have also been used to assess foetal health, however further investigation is warranted into the technique (Murchie et al., 2008).
All studies involved a multi-model approach to treatment with combined therapeutic regimes. 5/6 studies included altrenogest (Macpherson et  The use of antibiotics in a pulsatile manner has been explored in dogs with chronic pyoderma (Carlotti et al., 2004). In the canine model cephalexin (15 mg/kg twice a day) was prescribed for 2 days of the week and was compared against a placebo. The study found that pulse 'weekend' therapy was effective in reducing the time between relapses in canine idiopathic pyoderma and no resistance was noted in the 12 month period. It has been proposed that bacterial resistance to antibiotics may be limited in pulse antibiotic therapy as it minimises the time microbes are exposed to antibiotics and the selection of resistance (Baker et al., 2018). It was noted that in pulse therapy a bactericidal drug is advised with high concentrations to minimise pathogen abundance (Baker et al., 2018;and Carlotti et al., 2004). More research is required into what drugs and clinical scenarios pulse antibiotic therapy may be implemented.
Multiple types of placentitis have been described in the literature including ascending, focal mucoid (nocardioform), diffuse (haematogenous) and multifocal (Canisso et al., 2015). All the above studies involved an experimental model to induce ascending placentitis and the extrapolation to other forms is limited. Placentitis treatment remains frustrating for the clinician to treat with limited ability to perform culture and sensitivity against the potential organism/s involved. Current literature supports the use of antibiotics in the combined treatment of placentitis but does not provide evidence of the length of time they should be prescribed for. Further investigation of placentitis may involve correlating foetal and placental well-being to foal survival to be used as a measure of ceasing or altering treatment regimes.

Search Strategy
Databases searched and dates covered: Search terms were applied in PubMed Central accessed on NCBI website (1910-2019), CAB Abstracts database accessed on OVID platform (1973-2019)