Placentitis in Horses — a Major Cause of Late Term Foetal Loss

Home , Placentitis

Placentitis in Horses — A Major Cause of Late Term Foetal Loss —

RIRDC Publication No. 10/156

RIRDCInnovation for rural Australia

Placentitis in Horses

A Major Cause of Late Term Foetal Loss

by Joan Barbara Carrick

October 2010

RIRDC Publication No. 10/156 RIRDC Project No. PRJ-000744

ISBN 978-1-74254-075-7 ISSN 1440-6845

Placentitis in Horses: A Major Cause of Late-term Foetal Loss Publication No. 10/156 Project No. PRJ-000744

The information contained in this publication is intended for general use to assist public knowledge and discussion and to help improve the development of sustainable regions. You must not rely on any information contained in this publication without taking specialist advice relevant to your particular circumstances.

While reasonable care has been taken in preparing this publication to ensure that information is true and correct, the Commonwealth of Australia gives no assurance as to the accuracy of any information in this publication.

The Commonwealth of Australia, the Rural Industries Research and Development Corporation (RIRDC), the authors or contributors expressly disclaim, to the maximum extent permitted by law, all responsibility and liability to any person, arising directly or indirectly from any act or omission, or for any consequences of any such act or omission, made in reliance on the contents of this publication, whether or not caused by any negligence on the part of the Commonwealth of Australia, RIRDC, the authors or contributors.

The Commonwealth of Australia does not necessarily endorse the views in this publication.

This publication is copyright. Apart from any use as permitted under the Copyright Act 1968, all other rights are reserved. However, wide dissemination is encouraged. Requests and inquiries concerning reproduction and rights should be addressed to the RIRDC Publications Manager on phone 02 6271 4165.

Researcher Contact Details

Joan Barbara Carrick Scone Veterinary Hospital 106 Liverpool Street Scone NSW 2337

Phone: 02 6454 1333 Fax: 02 6545 2903 Email: [email protected]

In submitting this report, the researcher has agreed to RIRDC publishing this material in its edited form.

RIRDC Contact Details

Rural Industries Research and Development Corporation Level 2, 15 National Circuit BARTON ACT 2600

PO Box 4776 KINGSTON ACT 2604

Phone: 02 6271 4100 Fax: 02 6271 4199 Email: [email protected]. Web: http://www.rirdc.gov.au

Electronically published by RIRDC in October 2010 Print-on-demand by Union Offset Printing, Canberra at www.rirdc.gov.au or phone 1300 634 313

Foreword

Foetal losses and intensive care of critically ill neonates cost the horse industry an estimated $10 million to $20 million each year. Placentitis or inflammation of the placenta is most often caused by infection.

A major problem for horse breeders is that in many cases, placental pathology presents no clinical signs until the condition is so far advanced that effective treatment is difficult and frequently inadequate. Hence, placentitis is an extremely difficult condition to prevent and treat, primarily due to lack of knowledge about the development and progression of the condition. Improved understanding of the incidence and development of placentitis is critical, so that appropriate monitoring and prevention programs can be initiated in order to reduce the significant economic losses to the industry.

This study clearly demonstrates that detailed placental and foetal ultrasound examinations of broodmares in the field can be readily accomplished. Utilisation of the information collected to direct treatment of the mares resulted in an increase in the number of healthy foals produced by mares with a previous history of abortion or premature labor. This technology allows the early identification of brood mares that are at risk of pregnancy loss or delivery of a compromised foal so that early treatment can be initiated. In addition, the early initiation of treatment in this study dramatically improved neonatal foal health and survival. During the past three breeding seasons, many progressive thoroughbred breeders who participated in this research or were aware of the program have initiated proactive ultrasound monitoring of mares with a poor breeding history or of extremely high value. Widespread adoption of the program by horse breeders will significantly reduce reproductive wastage in the equine breeding industry.

This project was funded by industry contributions which were matched by funds from the Australian government through RIRDC.

This report is an addition to RIRDC’s diverse range of over 2000 research publications and it forms part of our Horse R&D program, which aims to assist in developing the Australian horse industry and enhancing its export potential.

Most of RIRDC’s publications are available for viewing, free downloading or purchasing online at www.rirdc.gov.au. Purchases can also be made by phoning 1300 634 313.

Craig Burns Managing Director Rural Industries Research and Development Corporation

iii

Acknowledgments

The assistance and support from owners, managers and staff at the thoroughbred brood mare farms involved in the project is gratefully acknowledged. In addition, I would like to acknowledge the invaluable assistance from the staff at Scone Veterinary Hospital, particularly, Dr Cameron Collins who convinced his clients to participate in the study, and Drs David O’Meara and Chris Quinn who examined numerous placentas. The assistance of Dr Angela Begg who assessed the histopathology of the placental samples is greatly appreciated. I would also like to acknowledge the assistance of Dr Arron Hodder and Dr Catherine Chicken who reviewed this report.

Abbreviations

CUPT Combined Utero-Placental Thickness cm Centimetres mm Millimetres ml Millilitres ng Nanograms mmol Milimoles

TMPS Potentiated Sulphonamide Antibiotic

Contents

Foreword ...... iii Acknowledgments...... iv Abbreviations ...... iv Executive Summary ...... vii Introduction ...... 1 Assessment and Treatment of the High Risk Pregnant Mare...... 2 Treatment ...... 3 Prevention Monitoring Program ...... 4 Objectives ...... 6 Specific Objectives ...... 6 Methodology ...... 7 Mare Selection ...... 7 Ultrasound Examination ...... 7 Foal Health and Placental Assessment ...... 9 Mare Treatment...... 9 Statistical Analysis ...... 9 Results ...... 11 Mare Treatment...... 11 Foal Health Outcomes from High Risk and Normal Mares ...... 11 Placental Pathology of High Risk and Normal Mares ...... 13 Placental Pathology and Neonatal Foal Health ...... 14 Ultrasound Parameters of High Risk and Normal Mares ...... 17 Ultrasound Parameters Associated with Gestational Age and Treatment ...... 19 Ultrasound Parameters Associated with Foal Health and Placental Pathology ...... 21 Incidence of Abnormal Scans and Placental Pathology ...... 23 Discussion ...... 26 Overall Objectives ...... 26 Specific Objectives ...... 26 Implications...... 28 Recommendations ...... 29 References ...... 30

Tables

Table 1 Foal health parameters in foals from high risk and normal mares...... 13 Table 2 The number of placentas with pathology present in the different regions of the placenta in normal and high risk mares...... 14

Table 3 Health parameters in foals from placentas with and without pathology...... 15

Table 4 Foal Health or Placental Pathology Variables that were correlated with ultrasound parameters have highlighted boxes with p values in the box ...... 22

Table 5 The percentage of ultrasound parameters that were abnormal in the entire study population of mares...... 24

Table 6 The percentage of ultrasound parameters that were abnormal in untreated mares that produced healthy foals...... 24

Table 7 Mean rectal and abdominal CUPT and median ultrasound score in normal mares that were not treated and produced a healthy foal...... 25

Figures

Figure 1 Mares were examined in the facilities used for normal reproductive examinations ...... 8 Figure 2 The foaling percentages for mares included in the study in the three years before the study during which they received no treatment, the two years during the study when they received treatment based on ultrasound parameters and the two years after the study, when only mares that received no ultrasound directed treatment were included ... 12 Figure 3 The white blood cell count in foals from mare that had pathology at the cervical pole of the placenta was lower than in foals from mare with no pathology at the cervical pole .. 16 Figure 4 Pathology present in the body of the placenta reduced the size of the foal and increased the time taken for the foal to stand...... 17 Figure 5 The rectal CUPT increased with gestational age similarly in both normal and high risk mares...... 18 Figure 6 The abdominal CUPT increased with gestational age similarly in both normal and high risk mares...... 18 Figure 7 The rectal and abdominal CUPT were very well correlated. Dotted lines represent 95% confidence interval...... 19 Figure 8 The foetal heart rate decreased with gestational age. Dotted lines indicate 95% confidence interval...... 20 Figure 9 The aortic diameter increased with gestational age. Dotted lines indicate 95% confidence interval...... 20 Figure 10 Mosaic plot of amniotic and allantoic fluid turbidity in treated compared with untreated mares...... 21 Figure 11 The derived model ultrasound index was correlated with the time taken for the foal to stand...... 23

Executive Summary What the report is about

This report describes a project that investigated the use of regular ultrasound examination of two groups of pregnant mares; one group had a normal foaling rate (normal) and the other group had repeatedly aborted or delivered premature foals (high risk). The mares were treated based on abnormalities identified during the ultrasound examinations. The regime of intensive examination and treatment resulted in a foaling rate of 90% for the high risk group. This was similar to the foaling rate for the normal mares. The project identified the ultrasound parameters that are important in the development of a high-risk pregnancy. Treatment of mares limited the validity of this aspect of the project, however an ultrasound index was developed. This index needs to be validated by use in a further prospective study.

Who is the report targeted at?

This report is aimed primarily at stud managers and veterinarians in the equine breeding industry. Application of the technology is expensive; however the successful delivery of healthy, valuable foals and reduced need for neonatal intensive care justifies the cost.

Background

Placentitis and placental abnormalities are a major cause of reproductive loss, which costs the equine industry a large amount of money each year due to lost foals and the treatment of critically ill foals. Because clinical signs of placental pathology are variable, typically the first indication of a problem is abortion or still birth. Due to undetected cases, the true incidence of placental pathology may be higher than current estimates. In addition, placental pathology contributes to development of critical illness in the equine neonate by predisposing to conditions such as prematurity, sepsis and perinatal asphyxia syndrome. Ultrasound examination of the foeto-placental unit can identify placental pathology before abortion or premature delivery occurs and is used to monitor clinical cases of mares with a systemic disease or precocious signs of imminent parturition to determine foetal health and development.

Aims/objectives

The primary objective of this study is to document the development and consequences of placental abnormalities occurring during the second half of gestation in thoroughbred broodmares and to determine whether treatment based on these findings improves foetal and neonatal foal health.

Methods used

During 2005 and 2006, ultrasound examinations were conducted every four weeks on 106 pregnant mares on three thoroughbred breeding farms in the Hunter Valley, NSW. The mares were treated with antibiotics and various combinations of anti-inflammatory drugs and altrenogest, a progesterone analogue. There were two groups of mares; a group with a normal breeding record and a group that had repeated abortions or delivery of a critically ill foal. Foaling of each mare was observed, the foal’s health recorded, a blood sample obtained and the placenta examined for abnormalities. All the ultrasound parameters were examined and found to have interaction with the gestational age at the time of data collection and with treatment of the mares; hence these two variables were included in all analysis. The ultrasound variables were used in a multivariable analysis of variance or logistic regression to determine association with foal health parameters and placental pathology. An ultrasound index was determined based on the parameters that had significant correlations.

vii

Results/key findings

Repeated ultrasound examinations and treatment of pregnant mares resulted in a foaling rate of 90% in the mares with a previous history of pregnancy loss which was the same as the foaling rate of the normal mares (91%). Foal health and placental pathology was correlated with placental width detected during both the trans-abdominal and trans-rectal examination, placental separation and folding detected during the trans-abdominal examination, foetal heart rate and foetal aortic diameter. An ultrasound index was derived using these parameters and found to be correlated with the time taken for the foal to stand.

Although there was correlation between foal health and placental pathology, the incidence of placental pathology (53%) was much higher than the incidence of critical illness in neonatal foals (9%). Intensive treatment of the mares during pregnancy may account for the observed improvement in neonatal foal health.

Implications for relevant stakeholders

Ultrasound examination and treatment of mares with a poor breeding history significantly improves the probability of delivery of a live, healthy foal. Adoption of this technology by the horse breeding industry and their veterinarians has the potential to reduce the losses in the breeding industry due to placental pathology.

Recommendations

Mares with a history of repeated abortion or delivery of a critically ill foal should, from mid- pregnancy through to delivery, have regular ultrasound examinations and treatment regimes should be based on these examinations.

Further research should be conducted using the ultrasound index derived in this project in a prospective study of normal brood mares that receive no treatment.

viii

Introduction

Each year approximately 20-30% of all mares bred in Australia fail to produce a live foal. Almost half of this failure is due to losses in the second half of gestation. Causes of pregnancy loss later in gestation may be due to foetal or placental anomalies, conception of twins, infection (viral, bacterial or fungal), hydrops amnion or allantois, loss of the structural integrity of the mare’s body wall or systemic illness in the mare.1, 2 Placental pathology is the primary contributing factor in the majority of cases of equine foetal loss.3, 4, 5, 6 In addition, placental pathology is a significant contributor to the delivery of premature, septic or dysmature foals that require expensive intensive care. The relative contribution of different placental pathologies to equine foetal and neonatal death varies with the definitions used by each investigator and by geographical location. However, placentitis is the single most common cause of placental pathology. Conception of twins has significantly reduced as a cause of abortion during the past 20 years 5, 6. This is most likely due to the increased use of ultrasound examination allowing early diagnosis. Umbilical elongation and torsion has become a more important cause of foetal and neonatal losses 5, 6. Placental pathology is frequently associated with umbilical cords longer than 85 cm and results in a reduction of the nutrient supply to the developing foetus. Limitation of nutrients reduces intrauterine growth rate of foals and frequently results in the presentation of a small for gestational age foal or in severe cases, abortion or stillbirth. In addition, regions of the placenta with reduced perfusion associated with an excessively long umbilical cord are more prone to infection and subsequent development of bacterial placentitis.

Typically placentitis is associated with bacterial or fungal infection which induces an acute or chronic response. The most common bacteria isolated from infected placentas are Streptococcus equi subspecies zooepidemicus, Escherichia coli, Klebsiella pneumonae and Pseudomonas sp.3, 5. Recently a model of ascending bacterial placentitis has been developed by a group of investigators at the University of Florida. This model is quite severe with the loss of approximately 90% of the foals. Not all mares exhibited clinical signs of infection nor gave indication of impending parturition, which is consistent with field cases of the condition7. The model has also provided valuable information about the pathophysiology of ascending placentitis and allowed investigation of some different treatment regimes8-12. The development of the ascending placentitis in this model was monitored using trans- rectal ultrasonography to measure the thickness of the combined utero-placental unit and a significant increase in thickness and separation of the utero-placental unit was detected by ultrasound examination 8.

The most common clinical presentation of placentitis is an ascending infection through the cervix1. The causes of a compromised cervical seal are varied. Stress to the foetus or mare can induce alteration of progesterone synthesis and metabolism which may result in abnormal relaxation of the cervix. It is apparently common for normal mares to have intermittent relaxation of the cervix and the specific factors that determine an individual mare’s risk of developing placentitis as a result of this relaxation are unknown. However the frequency and duration of relaxation of the cervix and the anterior vaginal micro flora may be significant factors. Bacterial vaginosis is an important risk factor in women at high risk for premature labour. Monitoring vaginal bacterial flora with early antibiotic treatment significantly improves the likelihood of a successful pregnancy in these women13. Once infection has been established in the placenta, the severity of the inflammatory response is critical. There is activation of the placental inflammatory cascade, primarily through resident macrophages, with synthesis of a range of cytokines and prostaglandins 14-16. These inflammatory mediators adversely affect the foetal metabolism resulting in significant foetal stress, induction of a foetal inflammatory response syndrome and enhanced myometrial contractility. Infection may be controlled by the use of antibiotics; however the foetus frequently remains severely compromised due to the inflammatory response. Improved identification and monitoring to allow early intervention is essential to improve outcomes for mares carrying a high risk pregnancy.

Assessment and Treatment of the High Risk Pregnant Mare.

History and Physical Examination

A detailed history of the mare’s previous pregnancies, previous parturitions, last date of service, the type of insemination, the number of services or inseminations required and the incidence of early embryonic loss is essential. A complete physical examination should be conducted on all mares carrying a high risk pregnancy. Any vaginal discharge, premature perineal relaxation or precocious mammary development is significant. If there is a vaginal discharge, some of the purulent material may be aseptically collected from the caudal vaginal without significantly disrupting the vestibular sphincter seal.

Examination of the udder and mammary secretions can provide useful information about the readiness of the foetus for extra uterine life. A sudden rise in milk calcium concentration greater than12 mmol/L indicates that the foetus is sufficiently mature for extra uterine life and that parturition is likely within 48 hours. Unfortunately the clinical signs of a high risk pregnancy can be very subtle and the first indication of a significant problem is acute abortion or impending premature labour. To significantly reduce foetal and neonatal losses it is essential to improve the clinical procedures used to assess high risk pregnant mares.

Hormonal Evaluation

Progesterone

The foeto-placental unit is the sole source of pregnanes and progestagens to maintain pregnancy from approximately 120 days through to parturition. The concentration of progesterone in the maternal circulation from mid-gestation to approximately 300 days of gestation is normally low. Commercial assays for progesterone measure variable amounts of the pregnanes and progestagens due to cross reactivity of the assay components which varies between different commercial assays. Placentitis induces a precocious rise in plasma progesterone levels but there is no clear link between plasma progesterone levels and placental pathology or foetal survival8, 17-19. The rise in maternal plasma progestagens may be due to dysregulation of progesterone synthesis in the abnormal placenta, excessive precursor secretion by the stressed foetal adrenal gland or a combination of both aetiologies. However, clinical signs of placentitis associated with low maternal progesterone levels, indicates a very poor prognosis for foetal survival and even with intensive monitoring and treatment of these cases the outcome is invariably disappointing.

Oestrogen

Between days 150 to 280 maternal concentration of oestrogen is highest and levels less than 500 ng/ml indicate severe foetal compromise and imminent abortion. After 280 days of gestation, interpretation of maternal oestrogen concentrations is limited and not well correlated with foetal survival. 19

Relaxin

From 70 days of gestation, relaxin is produced by the placenta and is detectable in the maternal circulation. There is an excellent correlation between low relaxin concentration and foetal-placental compromise, and maternal relaxin concentration could be used to monitor high risk pregnancies. Unfortunately validated assays of equine relaxin are not available commercially in Australia.

Ultrasound Examination of the Foetal-placental Unit

Ultrasound examination has been used to monitor late gestation pregnant mares with precocious signs of imminent parturition and the development of experimentally induced placentitis8, 20, 21, 22-25. Trans- rectal ultrasound examination allows investigation of the caudal section of the placenta, including the cervical pole region. This is the region through which most bacterial infections of the placenta occurs, hence is an excellent section of the placenta to monitor for ascending infections. After mid-gestation, trans-rectal ultrasound examination provides limited information about foetal development and health. Trans-abdominal ultrasound examination significantly increases the amount of the placenta that can be evaluated and allows monitoring of foetal activity, heart rate and growth rate. A biophysical score has been developed to assist in the prediction of foetal viability; however, the predictive accuracy of the published normal ultrasound parameters for foal survival has not been tested in a prospective clinical trial.

Repeated ultrasound assessment is also very useful for monitoring the response to treatment. Thickening of the combined utero-placental thickness (CUPT), degree of folding of the placenta and turbidity of the fluid all improve with successful treatment. Mares showing precocious signs of imminent parturition, that have no improvement in the ultrasonographic parameters after treatment, have a worse outcome than mares where improvement is recorded. (Carrick J., clinical experience)

Women with signs of preterm labour frequently undergo ultrasound assessment of the pregnancy. Several ultrasound parameters are associated with the risk of premature delivery and can be used to select women for referral to hospitals with tertiary level neonatal intensive care units prior to delivery26-29. There are numerous ongoing investigations into the use of ultrasound examination to evaluate women at risk for preterm labour so that treatment can be appropriately directed and preparations for the delivery of a compromised neonate can be initiated as required27.

Treatment

Mares with clinical signs of premature delivery that are not treated have a less than 30 % chance of delivering a live, healthy foal30, 31. In an experimental model of ascending placentitis where treatment was not provided, all the mares aborted or delivered a weak premature foal12. Treatment with antibiotics and anti-inflammatory drugs prolonged the gestation but did not improve foal viability. Addition of altrenogest to the treatment regime resulted in improved foetal viability and foal health1, 9, 10.

Antibiotics

The most common antibiotic class used to treat placentitis are potentiated sulphonamides (TMPS)2, 32. This antimicrobial group is broad spectrum, bactericidal and has excellent penetration of the placenta, readily reaching therapeutic concentrations in amniotic fluid and the foetus. Recently, penicillin and gentamicin have been shown to effectively penetrate the pregnant uterus and reach therapeutic levels in the allantoic fluid. This provides an excellent alternative treatment choice if there is a failure of response to TMPS.

Anti-inflammatory drugs

The inflammatory response to infection contributes significantly to foetal stress and predisposes the high risk pregnancy to premature parturition. There is significant synthesis of prostaglandins in response to infection and these substances increase myometrial contractility33, 34. It is important to control the inflammatory response and non-steroidal anti-inflammatory drugs should be administered to mares with a compromised pregnancy. Flunixin meglumine is very effective at inhibiting prostaglandin synthesis in response to many different inflammatory stimuli, particularly bacterial

toxins. The placental response to infection also results in significant synthesis of cytokines23, 12. Pentoxifylline (8.5mg/kg PO TID) is effective at reducing endotoxin induced cytokine synthesis by equine inflammatory cells35-37. In addition the drug may improve oxygenation of the placental by improving blood flow. There have been no controlled investigations into the efficacy of anti- inflammatory drugs in the treatment of placentitis.

Hormonal treatment

The use of progesterone or its synthetic analogue, altrenogest, in high risk pregnancy is widespread, however it is controversial. Administration of additional progesterone is counter-intuitive because maternal progesterone levels are already elevated and further increasing progestagen concentration is illogical. Progesterone and its metabolites possibly contribute to the successful treatment of a high risk pregnancy by maintenance of myometrial quiescence and reduction of the inflammatory response in the foetal-placental unit34, 38. The progesterone metabolite, 5 α pregnane inhibits the formation of oxytocin receptors, thus contributing to myometrial quiescence. Administration of a progesterone analogue, altrenogest, increased the level of 5α pregnane in the plasma of normal mares and in a separate study, prevented abortion of mid-term pregnancy in mares that were given a prostaglandin analogue18, 19, 39. As mentioned above, addition of altrenogest to the treatment regime of mares in an experimental model of ascending placentitis, greatly improved the number of foals that survived compared with a previous experiment using antibiotics and ant-inflammatory drugs alone. There is evidence that progesterone and its metabolites have significant anti-inflammatory effects and reduce the production of cytokine and prostaglandins by the inflamed placenta14, 33, 40, 41. In addition, meta- analysis of several prospective studies of women at high risk for preterm labour indicates that administration of progesterone significantly improves outcomes and prevents premature delivery, and recently the American Society of Obstetrics and Gynaecology has published the recommendation that progesterone should be administered to women with signs of impending preterm delivery that have a history of previous preterm labour42.

Ancillary treatment

Tocolytics – clenbuterol has been used to inhibit uterine contractility. Unfortunately, the doses required to reduce uterine contractility result in significant adverse side effects and the drug is not recommended2. Intranasal oxygen, to improve foetal oxygenation, Vitamin E as an antioxidant, and low dose aspirin, to improve placental oxygenation, have all been administered to mares at risk of premature labour. There is no data available to indicate the success of these treatments. Although aggressive treatment of the mare at high risk of premature parturition can improve foetal viability, frequently the foal that arrives is disappointing.

Prevention Monitoring Program

Because placentitis frequently presents no clinical signs until late in the development of the condition, a program to monitor mares with a poor breeding history is necessary to reduce foetal and neonatal losses. Mares that have repeatedly aborted or produced weak premature/dysmature or septic foals are at higher risk for foetal or neonatal loss compared to mares with a normal breeding history. Repeated ultrasound examinations at specified developmental milestones are a routine part of ante-natal care of women. Women who have had previous preterm deliveries are at high risk for preterm labour and poor foetal survival in subsequent pregnancies. Programs are being developed to identify the risk factors, monitor the women intensively and provide early intervention with antibiotics and progesterone. Early results from prospective studies using this type of management significantly improve outcomes in specific groups of women29, 43-46.

Until recently, ultrasound examination of mares with high risk pregnancies required admission to an equine hospital. This limited the number of examinations that were performed. In addition the detailed

ultrasound examination required to determine the biophysical profile took considerable time, further limiting its use in the field. New technology has allowed development of powerful portable ultrasound machines that are capable of excellent trans-abdominal examinations in the field. Thus it is now possible to monitor the progress of pregnancy in a large number of pregnant mares. In order to use this technology appropriately, it is essential that prospective clinical trials are undertaken to validate the ultrasound parameters used. Because there are significant losses associated with high risk pregnancies and treatment regimens are empirical, this study was conducted to determine whether ultrasound monitoring of equine pregnancies in the field is possible and if treatment based on the findings improved foetal and neonatal health.

Objectives

The primary objectives of this study are to document the development and consequences of placental abnormalities occurring during the second half of gestation in thoroughbred broodmares and to determine whether treatment based on these findings improves foetal and neonatal foal health.

Specific Objectives

• To correlate placental changes detected in mares by ultrasound, with gross and histopathological evidence of placentitis.

• To determine the incidence of placentitis in normal thoroughbred mares and in "high risk" mares, i.e. mares with histories of late term abortion, premature delivery or delivery of a foal with intrauterine growth retardation.

• To monitor the development and progression of placental abnormalities detected by ultrasound in normal and high risk thoroughbred mares.

• To determine whether equine foetal and neonatal weight and health are correlated with placental abnormalities detected by ultrasound.

Methodology

Mare Selection

The project was a prospective longitudinal observational study of thoroughbred broodmares. Pregnant thoroughbred mares resident at three breeding farms in the Hunter Valley were enrolled in the study. The mares (106) were selected by the stud managers, so that there were approximately equal numbers of mares with a normal breeding history and mares that were considered to be at high risk of late term abortion or delivery of a premature foal. The investigators were unaware of the classification of the individual mares during the study. Complete data was collected from 77 mares and used in the analysis of the results. The control group was 30 normally fertile thoroughbred mares (normal) and the study group was 47 mares (high risk) that have had repeated late term abortion, a stillbirth, a premature or small-for-gestational age septic foals. The project was conducted over two breeding seasons.

Ultrasound Examination

Each mare was examined by trans-rectal and trans-abdominal ultrasonography using a portable Terason 2000 ultrasound machine with a 5-7 MHz linear probe for the trans-rectal examinations and a 2-4 MHz curved array probe for the trans-abdominal examinations. The ventral abdomen was cleaned with a brush and soaked with methylated spirits. Mares were kept outdoors throughout winter, therefore the stud farm personnel requested that the ventral abdomen not be clipped. The first examination was conducted at 150-180 days of gestation and repeated at intervals of 28 days until the mare foaled.

Figure 1 Mares were examined in the facilities used for normal reproductive examinations

The CUPT was measured in three separate areas during the trans-rectal examination and recorded twice in three sites in the ventral midline of the abdomen as previously described20. The CUPT was graded 0 if normal and 1 if wider than normal. Turbidity of the foetal fluids was assessed as 0= normal, and 1= turbid. The placenta was examined for folding and separation from the endometrium and was recorded as 0 if normal and 1 if present for each parameter. The foetus was examined and

foetal activity, aortic diameter and heart rate were recorded. Examinations were conducted on the stud farms in the facilities used for normal reproductive examination of the mares and took approximately 20 minutes for each mare. Mares were not routinely sedated for the procedure and any mare that required repeated sedation was excluded from the study.

Foal Health and Placental Assessment

All aborted foetuses and dead foals were collected and a post mortem examination was conducted. Foals born alive were weighed 12-18 hours after birth, foal health recorded by the stud farm and blood samples obtained for a complete blood count, plasma fibrinogen, and serum creatinine concentrations. The exact length of gestation was recorded. Placentas from all mares in the study were examined by an equine veterinarian unaware of the group assignment of each mare and any pathology of the placenta was recorded. Two samples of the placenta, one adjacent to the cervical star and one from the body of the placenta close to the umbilical insertion, were obtained for histological examination. The histological examination was performed by a specialist pathologist who was unaware of the group assignment, treatment history or pregnancy outcome of the mare.

Mare Treatment

Due to the value of mares and foetuses in the study, all mares with abnormalities detected by ultrasound examination were treated with antibiotics and different combinations of anti-inflammatory drugs and altrenogest. Treatment of the mares was decided by the veterinarian responsible for the reproductive management of the farm. Because initiation of treatment was based on abnormal ultrasound parameters and was continued until the parameters became normal, associations of ultrasound parameters with neonatal foal health or with placental pathology were expected to be affected by treatment. Variability in the value and previous breeding history of the mares in this study and multiple veterinarians deciding treatment regimes, resulted in differences in timing of initiation and drug combinations used for each mare. In addition, the study was not designed to investigate treatment effects therefore precise details of treatments were not recorded and could not be analysed in detail.

Statistical Analysis

The data was entered from an excel spreadsheet to statistics program for analysis (JMP 7, SAS Institute Inc, Cary NC, USA). Descriptive analysis was calculated, data summarised and distribution of continuous data was evaluated for normality. Indicators of neonatal health were the length of gestation, death or admission to intensive care unit, foal’s complete blood count, neutrophil count, plasma fibrinogen, serum creatinine concentration, foal weight, placental weight, and time to stand and number of foals that took longer than 60 minutes to stand. Indicators of placental pathology were presence of pathology at any site in the foetal membranes; presence of pathology at the cervical star of the chorioallantois and presence of pathology at the body of the chorioallantois near the umbilical insertion: 0 = normal, 1= pathology and total pathology score.

The relationship between mare risk group and foal health parameters or placental pathologypatholgy was determined using Wilcoxin rank score tests. The relationship between gestational age at data collection or mare treatment with ultrasound parameters was tested by multivariable analysis of variance or logistic regression as appropriate. Means were compared using a Tukey-Kramer HSD, Wilcoxin Ranks Sum or Fishers Exact test. The relationship between ultrasound detected placental abnormality and neonatal foal health or placental pathology was determined using the dependent variables listed above respectively. The independent variables were gestational age at each ultrasound examination, mare treatment, and ultrasound parameters of rectal CUPT, rectal placental separation and folding, abdominal CUPT, abdominal placental separation and folding, foetal heart rate, foetal

aortic diameter, and fluid turbidity score. Forward stepwise regression was used to determine which parameters to include in the model. Gestational age at data collection and mare treatment was always included. A probability of 0.25 was used for inclusion. Associations between ultrasound parameters and foetal health or placental pathology were determined using multivariable analysis of variance or logistic regression as appropriate with a p< 0.05 considered significant. An ultrasound index was derived using the parameters that were found to have a significant association with foetal health or placental pathology.

Results

Mare Treatment

Mares in the high risk group were more likely to be treated with antibiotics but both groups received similar amounts of altrenogest. All of the mares in the high risk group and 27 of the 30 (90%) mares in the normal group were treated with antibiotics at least once during this study. The 47 high risk mares were examined 282 times during the study and it was decided to treat the mare after 80% of these examinations. The 30 mares in the normal group were examined 180 times during the study and it was decided to treat these mares after 57% of these examinations. In contrast, the expected incidence of placental abnormality in a group of normally fertile thoroughbred mares is approximately 10 to 20%3-5.. There is no previous data available about the incidence of placental abnormalities detected by ultrasound examination; however this project indicates that ultrasound examination detects abnormalities at a far higher incidence than the published rate of placental pathology. This may be due to an unexpectedly high rate of placental pathology in the normal group of mares during the study; however, after the study was completed and the mares returned to normal management, they were able to maintain normal pregnancy rates. This did not occur in those high risk mares that returned to normal management after the study was completed (see below). It is more likely that the ultrasound parameters that are currently used as normal values do not adequately discriminate a normal pregnancy from one that is at risk of premature delivery. Further research is necessary to better define parameters that can distinguish high risk pregnancies from normal.

Foal Health Outcomes from High Risk and Normal Mares

During the three years prior to the start of the study, high risk mares had a foaling rate of 38 % and normal mares had a foaling rate of 84 % (Figure 2). During this period the mares were not examined closely and received minimal treatment. However, during the study when mares were examined regularly and treated intensively, the high risk mares had a foaling rate of 90% and the normal mares had a foaling rate of 91%. After the study finished 16 mares from the high risk group were continued in a similar program and produced 27 foals from 31 pregnancies, 11 mares died or were retired and the remaining17 mares produced 13 foals from 32 pregnancies (41%) There were four mares from the normal group that were continued in a similar program and all produced foals each year, one mare from the normal group died and the remaining 25 mares produced 39 foals from 46 pregnancies in two years (82%) (Figure 2). This indicates that mares the stud managers identify as poor breeders have a lower foaling rate than mares identified as normal, and without intervention these mares will continue to have a lower foaling rate than normal mares. Ultrasound directed treatment of high risk mares from mid-gestation, before there are precocious signs of imminent parturition, produced the same foaling rate in these high risk mares as in normal mares. All mares in the study were treated based on the ultrasound parameters; an unintended consequence was that numerous mares in the normal breeding group with a normal reproductive history were treated.

100

90 27/30 42/47 Normal Mares 80 74/88 38/46 High Risk Mares 70

40 13/32 53/138 Foaling Percentage 30

0 2002, 2003 & 2004 2005 & 2006 2007 & 2008

Before Study Study Years After Study No Ultrasound Ultrasound No Ultrasound No Treatment and Treatment No Treatment

Figure 2 The foaling percentages for mares included in the study in the three years before the study during which they received no treatment, the two years during the study when they received treatment based on ultrasound parameters and the two years after the study, when only mares that received no ultrasound directed treatment were included. The numbers in the columns are the number of foals/number of pregnancies

The mares in the high risk group had a similar gestational length and their placental weights were similar to mares from the normal group. The foals from high risk mares weighed less, but took the same time to stand and suck as foals from normal mares. The percentage of foals that died or required intensive care were similar for the two groups of mares (Table 1). Foals from high risk mares had similar white blood cell count, neutrophil count, band neutrophil count and serum creatinine concentration but a lower fibrinogen than foals from normal mares. (Table 1)

High Risk Foal Health Parameter N Normal Mares N Mares 342 342 Gestational Length (days) 47 30 (328-365) (331-351) Percentage of foals that died or 10.00 47 12.77 23 required intensive care 53.1 56.0 * Foal Birth weight (kg) p=0.0421 38 23 (45-62) (50-64) 60 44 60 28 Time to Stand (min) (30-120) (25-90) 7.4 7.2 White Blood Cell Count x 109 42 25 (4.6-10.0) (5.4-9.6) 5.95 5.84 Neutrophil Count x 109 42 25 (3.1-7.8) (3.9-8.2) 0 0 Band Neutrophil Count x 109 42 25 (0-0.2) (0-0.5) 2 3 * Fibrinogen g/L p= 0.0399 42 25 (1-4) (2-3.4) 200 174 Creatinine mmol/L 32 21 (128-293) (122-258) 6.5 7.3 Placenta weight (kg) (4.9-8.1) 44 (5.0-8.5) 26

Table 1 Foal health parameters in foals from high risk and normal mares. Numbers are median with the numbers in parenthesis are the 90% quantile. The numbers in the “N” columns are the number of data points analysed These data indicate that ultrasound directed treatment of mares at high risk for foetal and neonatal loss reduces the number of pregnancies lost and the foals produced are similar to normal foals. This study has resulted in the use of ultrasound directed treatment in a large number of broodmares during the past three years. In 2007, yearlings from these mares sold at the two premier yearling sales in Australia had an average price that was 1.6 times higher than the overall sales average.

Placental Pathology of High Risk and Normal Mares

Forty-one of the 77 placentas examined had pathology in the placenta, detected either by gross or histological examination (Table 2). This is higher than would be expected based on the health of the foals. There was no difference in the overall incidence of pathology detected in the placenta between high risk and normal mares. Pathology in the body of the placenta was the only site that was associated with the risk group of the mare. High risk mares were more likely to have pathology in the body of the placenta than normal mares (p=0.156).

Gross Histo- Gross Histo- Gross Pathology Pathology pathology Pathology pathology Pathology at any Site at Cervical present at in Body of in Body of at other in the Pole Cervical Pole Placenta Placenta Sites Placenta

Total Mares 22 27 13 14 15 41 n=77 (28%) (34%) (16%) (17%) (18%) (53%) (%)

High Risk 14 17 11 11 10 26 Mares n=47 (30%) (36%) (23%) (23%) (21%) (55%)

Normal 7 9 1 2 4 15 Mares n=33 (23%) (30%) (3%) * (7%) * (13%) (50%)

Table 2 The number of placentas with pathology present in the different regions of the placenta in normal and high risk mares.

Placental Pathology and Neonatal Foal Health

The presence of pathology at any site in the placenta had no effect on the gestational length, size of the foal, whether the foal died or required intensive care, the white blood cell count, neutrophil count, band neutrophils count, serum fibrinogen level, serum creatinine concentration or the placental weight. The time taken for the foal to stand was the only parameter that was related to the presence of pathology at any site (Table 3).

Foal Health Parameters No Pathology N Pathology N 342 342 Gestational Length 36 41 (331-353) (328-366) Percentage of foals that died or 8% 36 15% 41 required intensive care 54 54 Birth Weight (kg) 27 34 (47-62) (42-64) 53 60* Time to Stand ( minutes) 34 38 (22—68) (30-120) 7.5 6.7 White Blood Cell Count x 109/L 32 35 (5.1-10.0) (4.4-9.7) 6 5.6 Neutrophil Count x 109/L 32 35 (3.9-8.0) (3.1-8.6) 2.5 2 Fibrinogen g/L 32 35 (1-4) (1-3) 184 196 Creatinine Umol/L 23 30 (137-280) (122-290) 6.5 7 Placental Weight ( kg) 33 37 (5-8.5) (4.9-8)

Table 3 Health parameters in foals from placentas with and without pathology. Numbers are medians with the 90% quantile in parenthesis, * indicates significantly different from foals produced by mares with placentas with no pathology.

Pathology at the cervical star reduced the white blood cell count and the neutrophil count (Figure3). The cervix is a common route of infection for the placenta and foetus. Infection of the neonate frequently is associated with reduced white cell count. Hence the association of reduced white cell count in foals from mares with cervical pole pathology may be due to a higher rate of infection in these foals. Unfortunately, there was no attempt to obtain blood cultures from the healthy foals or to isolate bacteria from the placentas in this study.

9 11 10 9 8 7 6 5

White BloodCell Count x 10 4 3 No Pathology Pathology Present

Cervical Pole of the Placenta

Figure 3 The white blood cell count in foals from mare that had pathology at the cervical pole of the placenta was lower than in foals from mare with no pathology at the cervical pole. The box plots of the two populations are shown, the width of the box plot represents the size of the population.

The presence of pathology in the body of the placenta was associated with reduced foal birth weight and increased time taken for the foal to stand (Figure 4). The size of the foal is dependant on the size of the placenta and the development of the chorionic surface.

70 160 65 140

60 e 120

t nu i

5 5 100 ) ( M

(K g

50 80 e and z S t

S i 45 60

o l t e oa

F 40 40

m i T 35 20 30 0 No Pathology Pathology No Pathology Pathology Present Present

Pathology in the Body of the Placenta

Figure 4 Pathology present in the body of the placenta reduced the size of the foal and increased the time taken for the foal to stand. Box plots of the two populations are shown, the width of the box plot represents the number in each population.

Ultrasound Parameters of High Risk and Normal Mares

Consistent with the foal health data and placental pathology data, there was no difference in ultrasound parameters, controlled for gestational age and treatment, between the high risk and normal mares.

Ultrasound parameters changed significantly with gestational age. Rectal CUPT and abdominal CUPT increased significantly as gestational age increased. The line of best fit for rectal CUPT was a polynomial 2nd degree line (Figure 5). Whereas the best fit for the abdominal CUPT and gestational age was linear (Figure 6). There was no difference in these parameters between the high risk mares and the normal mares.

1.5

1.3

1.1

0.9

0.7 Rectal (cm) CUPT 0.5

0.3

0.1 150 180 210 240 270 300 330 Gestational Age (days) High risk mares Normal Mares

Figure 5 The rectal CUPT increased with gestational age similarly in both normal and high risk mares. Dotted lines represent 95% confidence intervals.

1.7 1.5

1.3 1.1 0.9 0.7 Abdominal CUPT (cm) 0.5 0.3 0.1 150 180 210 240 270 300 330 Gestational Age (Days) High risk mares Normal Mares

Figure 6 The abdominal CUPT increased with gestational age similarly in both normal and high risk mares. Dotted lines represent 95% confidence interval.

The rectal and the abdominal CUPT were correlated with the best fit being linear. There was no difference between normal and high risk mares (Figure 7).

1.8

1.6

1.4 1.2 1 CUPT (cm) Abdominal 0.8 0.6 0.4 0.3 0.5 0.7 0.9 1.1 1.3 1.5 Rectal CUPT (cm) High risk mares Normal Mares

Figure 7 The rectal and abdominal CUPT were very well correlated. Dotted lines represent 95% confidence interval. These data indicate that high risk and normal mares have similar changes in their placentas during gestation and that when an abnormally wide placenta was detected by trans-rectal examination, there was a high probability that trans-abdominal ultrasound examination would detect the CUPT wider than normal.

Ultrasound Parameters Associated with Gestational Age and Treatment

Rectal and abdominal CUPT, and foetal aortic diameter increased and foetal heart rate decreased with gestational age (Figures 8&9). Foetal fluid turbidity did not change with gestational age. As gestational age increased, wider than normal CUPT values and placental separation detected by trans- rectal ultrasound examination were more likely to occur. As gestational age increased, a wider than normal CUPT and placental folding and separation detected during trans-abdominal ultrasound examination was more likely to occur. The rectal CUPT was correlated with the presence of separation and placental folding detected on the trans-rectal ultrasound. The abdominal CUPT was correlated with the presence of separation and placental folding detected on the trans-abdominal ultrasound. These findings indicate although there are changes in the ultrasound parameters with gestational age, that there is close correlation in abnormal ultrasound parameters, and when one parameter is abnormal other parameters are typically also abnormal.

160

140

120

100

80 Foetal Heart Rate (bpm) 60 150 180 210 240 270 300 330

Gestational Age (days)

Figure 8 The foetal heart rate decreased with gestational age. Dotted lines indicate 95% confidence interval.

2.8

2.4

1.6

1.2

0.8 Foetal Aortic Diameter (cm) Diameter Aortic Foetal 150 180 210 240 270 300 330 Gestational Age (days)

Figure 9 The aortic diameter increased with gestational age. Dotted lines indicate 95% confidence interval.

Rectal and abdominal CUPT and foetal aortic diameter were reduced and foetal heart rate was increased in mares that were treated. Foetal fluids were more likely to be turbid in mares that were treated (Figure 9). This result would be expected as mares were treated based on abnormal ultrasound findings. Placental folding was more likely to be detected during trans-rectal and trans-abdominal ultrasound in mares that were treated. However, placental separation detected during trans-rectal and trans-abdominal ultrasound was not associated with the treatment of the mare.

1.00

Turbid 0.75

0.50

Fluid Turbidity Fluid Clear 0.25

0.00 Not Treated Treated Mares

Figure 10 Mosaic plot of amniotic and allantoic fluid turbidity in treated compared with untreated mares.

Ultrasound Parameters Associated with Foal Health and Placental Pathology

The mares were treated based on the ultrasound findings which resulted in an increase in foetal and neonatal survival (Figure 2). The remarkably low incidence of disease in the neonatal foals in this study restricted detection of any association between foal health and ultrasound parameters. Gestational length decreased in association with wider than normal rectal CUPT and abdominal placental separation. The foal’s size was correlated with the aortic diameter and the foetal heart rate. The number of foals that died or require intensive care was associated with the abdominal CUPT. An increased time for the foal to stand was associated with abdominal placental separation and folding and the foetal heart rate. The foal’s white blood cell count, neutrophil count, fibrinogen and creatinine concentration and the placental weight were not correlated with any of the ultrasound parameters (Table 4).

Placental pathology was associated with wider than normal CUPT, separation and folding detected during the trans-abdominal examination only. No other ultrasound parameters were associated with placental pathology. Further prospective investigation of ultrasound assessment of pregnant mares needs to be completed to be able to determine the important discriminatory ultrasound parameters for foetal survival and placental pathology (Table 4).

Placental parameters Placental parameters Foal Health detected during Rectal detected during Abdominal or Placental Foetal Foetal Aortic Exam Exam Fluid Pathology Heart Rate Diameter Parameter CUPT Separation Folding CUPT Separation Folding

Gestational Length p=.0003 p=.0157 (days) Percentage of foals that died or p=.0493 required intensive care

Foal Birth p=.0232 p= .009 weight (kg)

Time to p=.0195 p=.0003 p=.0007 Stand (min)

White Blood Cell Count x 109

Neutrophil

Count x 109

Band Neutrophil Count x 109

Fibrinogen g/L

Creatinine mmol/L

Placenta p=.0227 p=.0499 weight (kg)

Pathology at Cervical Pole Pathology in Body of p=.0147 Placenta

Pathology at any Site in the Placenta

Table 4 Foal Health or Placental Pathology Variables that were correlated with ultrasound parameters have highlighted boxes with p values in the box

Ultrasound parameters that were correlated with foetal health or placental pathology (rectal and abdominal CUPT, placental separation and folding detected during the trans-abdominal ultrasound examination, foetal aortic diameter and foetal heart rate) were used to derive a model ultrasound index. This ultrasound index was positively correlated with the foal’s time to stand and with presence of pathology in the placenta (Figure 11). Further prospective investigation of ultrasound assessment of pregnant mares is necessary to determine the important discriminatory and ultrasound parameters for foetal survival.

4 log time to stand to time log

0 1 2 3 4 5 Ultrasound Index

Figure 11 The derived model ultrasound index was correlated with the time taken for the foal to stand. Box plots for the time to stand at each index level are represented.

Incidence of Abnormal Scans and Placental Pathology

An abnormal ultrasound parameter was detected at 92% of examinations. The percentage of abnormal ultrasound parameters did not vary with the origin of the ultrasound image (trans-rectal or trans- abdominal) or with the risk group of the mare (Table 5). This incidence is similar to that expected for high risk mares but much higher than expected for normal mares and consequently, almost all the mares were treated at least once during the study (see Results: Mare Treatment). These data indicate that the current normal values, when used prospectively, over-estimate the incidence of placental pathology in a population of mares. The created model ultrasound index was four or greater in only 8 % of examinations; in addition, utilisation of this parameter would have resulted in the treatment of 31 % of the mares. Further prospective clinical trials need to be conducted on mares that are not treated so that the predictive value of ultrasound parameters to determine equine pregnancy outcome can be accurately evaluated and excessive treatment avoided.

Rectal Abdominal CUPT Rectal Rectal Abdominal Abdominal Model CUPT wider Placental Placental Placental Placental Ultrasound wider than than Separation Folding separation Folding Score >4 normal All Mares normal Percentage of 46% 20% 38% 46% 36% 37% 17% Ultrasound Parameters

Table 5 The percentage of ultrasound parameters that were abnormal in the entire study population of mares.

When the data from mares that received no treatment and produced a healthy foal was summarised, the number of abnormal observations was reduced, but remained about 30%. (Table 6) However, only 3% of the examinations of these mares resulted in a total ultrasound score of 4 or greater, suggesting that the derived ultrasound score maybe a better discriminatory variable than individual ultrasound parameters for foetal survival and health. There were no mares in the study that were not treated and aborted or delivered a critically ill foal, so comparisons based on foal outcome from untreated mares could not be performed.

Rectal Untreated CUPT Rectal Rectal Abdominal Abdominal Abdominal Model Mares wider Placental Placental CUPT wider Placental Placental Ultrasound Healthy than Separation Folding than normal Separation Folding Score >4 Foals normal Percentage of 37% 23% 28% 49% 23% 21% 3% Ultrasound Parameters

Table 6 The percentage of ultrasound parameters that were abnormal in untreated mares that produced healthy foals.

The mean rectal and abdominal CUPT from mares that produced a normal foal and received no treatment at each gestational age was similar to previously published data 20. In this group of mares the maximum ultrasound score exceeded 4 on only 5 occasions and the median total ultrasound score was 2 or less at all gestational ages examined (Table 7).

Std Dev Mean Std Dev Median Gestational Mean Rectal (Rectal Abdominal (Abdominal Ultrasound Age CUPT (cm) CUPT) CUPT (cm) CUPT) Score

150-180 0.48 0.09 0.64 0.13 2 180-210 0.49 0.08 0.69 0.15 1 210-240 0.55 0.11 0.74 0.14 .5 240-270 0.66 0.10 0.87 0.14 1 270-300 0.79 0.10 0.98 0.11 2 300-330 0.98 0.17 1.06 0.13 2

Table 7 Mean rectal and abdominal CUPT and median ultrasound score in normal mares that were not treated and produced a healthy foal.

This study has generated considerable clinical work examining pregnant mares that the owners value highly. Unfortunately the current techniques have not been validated in a thorough prospective trial using normal mares. This study shows clearly that the outcome of high risk pregnancies can be improved using this technology, however an unacceptably high number of normally fertile mares were treated because of the high rate that abnormalities were detected. One of the secondary aims of this study was to reduce the indiscriminate use of antibiotics in broodmares; unfortunately this technology has the opposite effect. It is critically important to conduct a prospective trial using these parameters, particularly the total ultrasound score, in a large group of normal mares that are not treated so that the positive and negative predictive values can be determined.

Discussion

Overall Objectives

The primary objective, to document the development and consequences of placental abnormalities occurring during the second half of gestation in thoroughbred broodmares and determine whether treatment based on these findings improves foetal and neonatal foal health was achieved. It is possible to conduct a modified detailed placental and foetal ultrasound examination of broodmares in the field. Utilisation of the data collected to direct treatment of the mares resulted in an increase in the number of healthy foals produced by mares with a previous history of abortion or premature labor. The technology detects subtle changes in the placenta and results in a large number of mares receiving treatment during a monitored pregnancy. It is expected that many of the normally fertile mares would control the mild placental pathology detected and deliver a normal foal without treatment. Therefore, it is critically important that improved discriminatory parameters are developed so that treatment can be more efficiently directed.

Specific Objectives

To correlate placental changes detected in mares by ultrasound, with gross and histopathological evidence of placentitis.

To determine whether equine foetal and neonatal weight and health are correlated with placental abnormalities detected by ultrasound.

These two objectives could not be adequately achieved because treatment of the mares was more successful than expected. Treatment of the mares was determined by the ultrasound detected changes in the placenta and significantly affected these parameters. In addition, treatment improved foetal and neonatal foal health; hence there were significantly fewer cases of foetal loss or critically ill neonates than expected. Consequently, the data available to determine correlations between ultrasound detected placental abnormalities and foal health or placental pathology was insufficient for adequate statistical analysis. The significant disadvantage of this failure to meet these objectives is that numerous valuable mares now have regular ultrasound assessments of their pregnancies and a large number of mares receive treatment that may be unnecessary because pregnancies with a high probability of a poor outcome cannot be accurately discriminated from those that will have a normal outcome. Further work to improve the predictive values of ultrasound examination is essential.

To monitor the development and progression of placental abnormalities detected by ultrasound in normal and high risk thoroughbred mares.

This objective was limited by the treatment of mares during the study, however abnormalities were detected during both trans-rectal and trans-abdominal examination and changes in these abnormalities over time were observed. When treatment was initiated, the changes in the ultrasound parameters typically returned to normal. In many mares when treatment was discontinued, the ultrasound parameters increased again to be outside the normal ranges. This was more noticeable in high risk mares. Because the study was not designed to assess the effect of treatment, precise records of treatment regimes are not available and further analysis of the changes noticed within mares as a response to treatment unfortunately cannot be performed.

To determine the incidence of placental pathology in normal thoroughbred mares and in mares at high risk for abortion or premature delivery.

Pathology was detected in the placenta of 53% of all the mares examined, although fewer normal mares (47%) had some pathology detected in their placentas than high risk mares (55%); this difference was not significant. These data are similar to previous reports from thoroughbred mares where histopathology was found in 23/50 (46%) of the placentas examined 30. High risk mares were more likely to have pathology in the body of the placenta than normal mares. The survival of foals from placentas with histological abnormalities has previous been reported to be less than 10% (2/23) (30). In contrast, in this study survival of foals from placentas with histological abnormalities was greater than 80% (24/29). The histopathology in the placenta of mares in this study was assessed by an experienced equine specialist pathologist completely unaware of the mare’s risk, treatment regime or pregnancy outcome. Both studies included examination of similar numbers of placentas. It is possible although pathology developed in the placenta, that treatment during pregnancy improved foetal health and neonatal survival. This result is particularly encouraging for the horse breeding industry; because it indicates that intensive monitoring and ultrasound directed treatment results in the delivery of healthy foals from mares with suboptimal placental development. Assessment of the racing performance of the foals from this study is currently underway and will be the final determinant of the success of this intensive intervention.

Implications

The group of 47 high risk mares produced 43 healthy foals during this study. Based on previous foaling records and foaling records in the two years subsequent to the study, these mares would normally be expected to produce only19 foals. Adoption of this technology should be able to halve the number of pregnancies lost after conception which could improve the average annual foaling rate by up to 10%. Importantly, the size and health of these foals was similar to foals from the normal mares. During the 2006, 2007 and 2008 breeding seasons, numerous thoroughbred breeding farms in the Hunter Valley and in the Hawkesbury region have adopted the program to improve the productivity of their mares. The program has been so successful in the region that there are now several other veterinarians using the technology to monitor high risk pregnancies. In addition, the owners of numerous valuable mares with normal breeding records have requested ultrasound monitoring during pregnancy. The use of this technology in normal mares resulted in a large number receiving unnecessary treatment; it is critically important that the ultrasound index derived in this project is investigated in a prospective study in a large group of normal mares that receive no treatment.

Recommendations

This study has clearly demonstrated that ultrasound monitoring and treatment of high risk mares improves foetal and neonatal viability. It is important that the technology is further validated by prospective studies in untreated mares and refined to be more useful in evaluating normal equine pregnancies. Preventive ante-natal monitoring programs similar to those used in women could then be developed. Information derived from this study needs to be disseminated to equine veterinarians throughout Australia and internationally by presentations at veterinary meetings, workshops and publications in veterinary journals.

References

1. Macpherson ML. Diagnosis and treatment of equine placentitis. Vet Clin North Am Equine Pract. Dec 2006;22(3):763-776. 2. Macpherson ML, Bailey CS. A clinical approach to managing the mare with placentitis. Theriogenology. Aug 2008;70(3):435-440. 3. Giles RC, Donahue JM, Hong CB, et al. Causes of abortion, stillbirth, and perinatal death in horses: 3,527 cases (1986-1991). J Am Vet Med Assoc. Oct 15 1993;203(8):1170-1175. 4. Hong CB, Donahue JM, Giles RC, Jr., et al. Equine abortion and stillbirth in central Kentucky during 1988 and 1989 foaling seasons. J Vet Diagn Invest. Oct 1993;5(4):560-566. 5. Smith KC, Blunden AS, Whitwell KE, Dunn KA, Wales AD. A survey of equine abortion, stillbirth and neonatal death in the UK from 1988 to 1997. Equine Vet J. Jul 2003;35(5):496- 501. 6. Ricketts SW, Barrelet A, Whitwell KE. Equine Abortion. Equine Vet Educ, Manual. 2003;6:18-21. 7. CalderwoodMays MB, LeBlanc MM, Paccamonti DL. Route of fetal infection in a model of ascending placentitis. Theriogenology. 2002;58:791-792. 8. Morris S, Kelleman AA, Stawicki RJ, et al. Transrectal ultrasonography and plasma progestin profiles identifies feto-placental compromise in mares with experimentally induced placentitis. Theriogenology. Mar 1 2007;67(4):681-691. 9. Bailey CS, MacPherson ML, Graczyk j, et al. Treatment efficacy of trimethorpim sulphamtheoxazole, pentoxifylline and altrenogest in equine placentitis. Theriogenology. 2007;68:516-517. 10. Murchie TA, Macpherson ML, LeBlanc MM, Luznar S, Vickroy TW. Continuous monitoring of penicillin G and gentamicin in allantoic fluid of pregnant pony mares by in vivo microdialysis. Equine Vet J. Nov 2006;38(6):520-525. 11. McGlothlin JA, Lester GD, Hansen PJ, et al. Alteration in uterine contractility in mares with experimentally induced placentitis. Reproduction. Jan 2004;127(1):57-66. 12. LeBlanc MM, Giguere S, Brauer K, et al. Premature delivery in ascending placentitis is

associated with expression of cyokines an allantoic fluid prostaglandins E2 and F2α. Theriogenology. 2002;58:841-844. 13. McDonald HM, Brocklehurst P, Gordon A. Antibiotics for treating bacterial vaginosis in pregnancy. Cochrane Database Syst Rev. 2007(1):CD000262. 14. Romero R, Espinoza J, Kusanovic JP, et al. The preterm parturition syndrome. BJOG. Dec 2006;113 Suppl 3:17-42. 15. Gotsch F, Romero R, Kusanovic JP, et al. The fetal inflammatory response syndrome. Clin Obstet Gynecol. Sep 2007;50(3):652-683. 16. Blank V, Hirsch E, Challis JR, Romero R, Lye SJ. Cytokine signaling, inflammation, innate immunity and preterm labour - a workshop report. Placenta. Mar 2008;29 Suppl A:S102-104. 17. Santschi EM, LeBlanc MM, Weston PG. Progestagen, oestrone sulphate and cortisol concentrations in pregnant mares during medical and surgical disease. J Reprod Fertil Suppl. 1991;44:627-634. 18. Ousey JC, Houghton E, Grainger L, Rossdale PD, Fowden AL. Progestagen profiles during the last trimester of gestation in Thoroughbred mares with normal or compromised pregnancies. Theriogenology. Apr 15 2005;63(7):1844-1856. 19. Ousey JC. Hormone profiles and treatments in the late pregnant mare. Vet Clin North Am Equine Pract. Dec 2006;22(3):727-747. 20. Bucca S, Fogarty U, Collins A, Small V. Assessment of feto-placental well-being in the mare from mid-gestation to term: transrectal and transabdominal ultrasonographic features. Theriogenology. Aug 2005;64(3):542-557.

21. Reimer JM. Use of transcutaneous ultrasonography in complicated latter-middle to late gesatioan pregnancies in the mare : 122 case. Proceedings American Association of Equine Practioners Meeting. 1997;43:259-261. 22. Reef VB, Vaala WE, Worth LT, Sertich PP, Spencer P. Ultrasonographic assessment of fetal well-being durnig late gestation: development of an equine biophysical profile. Equine Veterinary Jounral. 1996;28(3):200-208. 23. Reef VB, Vaala WE, Worth LT, Spencer P, Hammett B. Ultrasonographic evaluation of the fetus and intrauterine environment in healthy mares during late gestation. Veterinary Radiology and Ultrasound. 1995;36(6):533-541. 24. Troedsson MH, C.D. R, Zent WW, Steiner JV. Transrectal ultrasonography of hte placenta in normal mare and mare with pending abortion : A field study. Proceedings American Association of Equine Practioners Meeting. 1997;43:256-258. 25. Renaudin CD, Troedsson MH, Gillis CL, King VL, Bodena A. Ultrasonographic evaluation of the equine placenta by transrectal and transabdominal approach in the normal pregnant mare. Theriogenology. 1997;47:559-573. 26. Alfirevic Z, Allen-Coward H, Molina F, Vinuesa CP, Nicolaides K. Targeted therapy for threatened preterm labor based on sonographic measurement of the cervical length: a randomized controlled trial. Ultrasound Obstet Gynecol. Jan 2007;29(1):47-50. 27. Santos IS, Matijasevich A, Silveira MF, et al. Associated factors and consequences of late preterm births: results from the 2004 Pelotas birth cohort. Paediatr Perinat Epidemiol. Jul 2008;22(4):350-359. 28. Celik E, To M, Gajewska K, Smith GC, Nicolaides KH. Cervical length and obstetric history predict spontaneous preterm birth: development and validation of a model to provide individualized risk assessment. Ultrasound Obstet Gynecol. May 2008;31(5):549-554. 29. Crane JM, Hutchens D. Transvaginal sonographic measurement of cervical length to predict preterm birth in asymptomatic women at increased risk: a systematic review. Ultrasound Obstet Gynecol. May 2008;31(5):579-587. 30. Cottrill CM, Jeffers-Lo J, Ousey JC, et al. The placenta as a determinant of fetal well-being in normal and abnormal equine pregnancies. J Reprod Fertil Suppl. 1991;44:591-601. 31. Rossdale PD, Ousey JC, Cottrill CM, Chavatte P, Allen WR, McGladdery AJ. Effects of placental pathology on maternal plasma progestagen and mammary secretion calcium concentrations and on neonatal adrenocortical function in the horse. J Reprod Fertil Suppl. 1991;44:579-590. 32. LeBlanc MM, Macpherson ML, Sheerin PC. Ascending Placentitis:What we know about pathophysiology, diagnosis and treatment. Proceedings American Accosication of Equine Practioners Meeting. 2004;50:1417-1430. 33. Raghupathy R. Manipulation of cytokine production profiles as a therapeutic approach for immunologic pregnancy loss. Indian J Biochem Biophys. Aug 2008;45(4):229-236. 34. Mesiano S. Myometrial progesterone responsiveness. Semin Reprod Med. Jan 2007;25(1):5- 13. 35. Barton MH, Ferguson D, Davis PJ, Moore JN. The effects of pentoxifylline infusion on plasma 6-keto-prostaglandin F1 alpha and ex vivo endotoxin-induced tumour necrosis factor activity in horses. J Vet Pharmacol Ther. Dec 1997;20(6):487-492. 36. Barton MH, Moore JN, Norton N. Effects of pentoxifylline infusion on response of horses to in vivo challenge exposure with endotoxin. Am J Vet Res. Nov 1997;58(11):1300-1307. 37. Baskett A, Barton MH, Norton N, Anders B, Moore JN. Effect of pentoxifylline, flunixin meglumine, and their combination on a model of endotoxemia in horses. Am J Vet Res. Nov 1997;58(11):1291-1299. 38. Sfakianaki AK, Norwitz ER. Mechanisms of progesterone action in inhibiting prematurity. J Matern Fetal Neonatal Med. Dec 2006;19(12):763-772. 39. Ousey JC. Peripartal endocrinology in the mare and foetus. Reprod Domest Anim. Aug 2004;39(4):222-231.

40. Elovitz MA, Gonzalez J. Medroxyprogesterone acetate modulates the immune response in the uterus, cervix and placenta in a mouse model of preterm birth. J Matern Fetal Neonatal Med. Apr 2008;21(4):223-230. 41. Raghupathy R, Al Mutawa E, Makhseed M, Al-Azemi M, Azizieh F. Redirection of cytokine production by lymphocytes from women with pre-term delivery by dydrogesterone. Am J Reprod Immunol. Jul 2007;58(1):31-38. 42. ACOG Committee Opinion number 419 October 2008 (replaces no. 291, November 2003). Use of progesterone to reduce preterm birth. Obstet Gynecol. Oct 2008;112(4):963-965. 43. Morgan MA, Goldenberg RL, Schulkin J. Obstetrician-gynecologists' screening and management of preterm birth. Obstet Gynecol. Jul 2008;112(1):35-41. 44. Fahey JO. Clinical management of intra-amniotic infection and chorioamnionitis: a review of the literature. J Midwifery Womens Health. May-Jun 2008;53(3):227-235. 45. Nygren P, Fu R, Freeman M, Bougatsos C, Klebanoff M, Guise JM. Evidence on the benefits and harms of screening and treating pregnant women who are asymptomatic for bacterial vaginosis: an update review for the U.S. Preventive Services Task Force. Ann Intern Med. Feb 5 2008;148(3):220-233. 46. Mari G, Hanif F, Treadwell MC, Kruger M. Gestational age at delivery and Doppler waveforms in very preterm intrauterine growth-restricted fetuses as predictors of perinatal mortality. J Ultrasound Med. May 2007;26(5):555-559; quiz 560-552. 47. Axon JE, Palmer JE. Clinical pathology of the foal. Vet Clin North Am Equine Pract. 2008;24(2):357-386.

Placentitis in Horses by Joan Barbara Carrick Publication No. 10/156

Placentitis or inflammation of the placenta is most often RIRDC is a partnership between government and industry caused by infection. Placentitis and placental abnormalities to invest in R&D for more productive and sustainable rural are a major cause of reproductive loss, which costs the equine industries. We invest in new and emerging rural industries, a industry a large amount of money each year due to lost foals suite of established rural industries and national rural issues. and the treatment of critically ill foals. Because clinical signs of placental pathology are variable, typically the first indication of a Most of the information we produce can be downloaded for free problem is abortion or still birth. or purchased from our website .

Ultrasound examination can identify placental pathology before RIRDC books can also be purchased by phoning abortion or premature delivery occurs. In this study, mares were 1300 634 313 for a local call fee. treated based on abnormalities identified during the ultrasound examinations. The regime of intensive examination and treatment resulted in a foaling rate that was similar to the foaling rate for normal mares.

Cover photo: Ultrasound examination on a pregnant mare to detect Placentitis and placental abnormalities

Contact RIRDC: Level 2 15 National Circuit Ph: 02 6271 4100 Most RIRDC publications can be viewed and purchased at Barton ACT 2600 Fax: 02 6271 4199 our website: Email: [email protected] PO Box 4776 web: www.rirdc.gov.au www.rirdc.gov.au Kingston ACT 2604 Bookshop: 1300 634 313

RIRDCInnovation for rural Australia