Article Text

Two cases of equine pregnancy loss associated with Leptospira infection in England
  1. K. E. Whitwell, BVSc, DiplECVP, FRCVS1,
  2. A. S. Blunden, FRCPath, PhD, MRCVS1,
  3. J. Miller, BSC1 and
  4. J. Errington, BSc, MSc2
  1. 1 Animal Health Trust, Lanwades Park, Kentford, Newmarket, Suffolk CB8 7UU
  2. 2 Veterinary Laboratories Agency ‐ Penrith, Merrythought, Calthwaite, Penrith, Cumbria CA11 9RR
  1. E-mail for correspondence:

Statistics from

INFECTION with Leptospira can cause abortions and stillbirths in domestic cattle, sheep and pigs, as well as in wild animals. It has also been well documented as a cause of abortions in mares in several countries, including the USA, Canada, South America, New Zealand, Australia and some European countries. However, apart from Northern Ireland (Ellis and others 1983, Ellis and O'Brien 1988), leptospirosis has not been positively identified as a cause of equine pregnancy loss in the UK. This short communication describes two cases of pregnancy loss associated with Leptospira infection, which are believed to be the first documented cases in England.

Horses frequently have antibodies to several Leptospira serovars, indicating that they have been exposed, but not necessarily that they are clinically infected (Bernard 1993). In the UK, the most common serovars are Bratislava, Icterohaemorrhagiae, Canicola and Hardjo (Hathaway and others 1981), in many countries, serovars Pomona and Grippotyphosa are also common (Szeredi and Haake 2006).

In 2007, 42 samples of equine fetal kidney (cortex and medulla) were collected during diagnostic postmortem examinations, placed in transport medium and submitted for Leptospira PCR analysis to Veterinary Laboratories Agency — Weybridge. The routine investigations on the fetus and placenta included gross dissection, microbiology and histopathology. Immunohistochemistry for Leptospira was carried out retrospectively on two cases, one that was Leptospira PCR positive and showed unusual fetal lesions (case 1), and one that showed unusual placental lesions (case 2).

For PCR testing, kidney tissue was macerated and boiled, and the DNA was extracted using a Roche MagNA Pure LC instrument. The Leptospira PCR, which was performed on a Roche LightCycler, employs two separate reactions, one to detect DNA from pathogenic leptospires, and the other to check for inhibition of the PCR. Fluorescent probes detect the production of amplicon throughout the PCR in real time. Positive samples produce an amplification curve. The point at which the measured fluorescence increases above a calculated background level is called the cycle threshold (Ct) value and gives an indication of the amount of template Leptospira DNA present in the sample. The PCR amplicon was prepared for DNA sequencing by reamplifying it with nested primers. The PCR product was cleaned up by gel electrophoresis and submitted for DNA sequencing. The sequence data were analysed using Lasergene 6 and compared with genomes in the National Centre for Biotechnology Information (NCBI) database.

Immunohistochemistry was performed on two sets of formalin-fixed, paraffin wax-embedded tissues. Sections were cut at 3 μm on to positively charged 75 μm capillary gap slides and incubated overnight at 56°C. Rehydrated sections underwent a high pH citrate retrieval (S3307; Dako REAL) and were stained using an automated two-layered indirect technique. Sections were counterstained with haematoxylin (S2020; Dako REAL). Endogenous activity was blocked with a peroxidase solution and the target enzyme reaction was developed with diaminobenzidine and chromogen. Primary polyclonal Leptospira antiserum (ARA-050; National Veterinary Services Laboratories), was diluted 1:400 in antibody diluent (S2022; Dako REAL) and run alongside negative control sections. The conjugated secondary antibody was pig anti-rabbit (P0217; Dako REAL) diluted 1:100 in a buffered solution. Immunohistochemically confirmed Leptospira-positive equine tissue was used as a positive control.

In case 1, a five-year-old primigravid thoroughbred mare, which had been vaccinated against equine herpesvirus type 1 (EHV-1), gave birth to a stillborn foal at 324 days of gestation after showing no sign of impending parturition. Examination revealed a well-developed foal weighing 51 kg with no congenital defects. Evidence indicated that the foal was alive during second-stage labour but suffered terminal asphyxia and died before the umbilical cord had ruptured. There were recent cardiac haemorrhages and the brain was intensely congested, but the limbs were straight and there was no sign of any severe thoracic trauma or haemarthrosis. The chorion was everted and of normal shape. Most of the non-pregnant horn was missing. The cervical star had ruptured normally and there was no thickening or discoloration suggestive of placentitis. However, there was some localised oedema, congestion and mild recent haemorrhage in the body rostral to the cervical star. An apparently incidental finding was an unusually long (100 cm), narrow allantoic cyst in the rostral ventral body, coated by hippomane debris. The amnion and umbilical cord appeared normal. No bacteria were isolated in aerobic culture from liver or lung swabs. Streptococcus zooepidemicus, Escherichia coli, Acinetobacter lwoffii, a coagulase-negative staphylococcus and an a-haemolytic streptococcus were isolated from the cervical star. PCR analysis of foal and placental tissues proved negative for EHV-1 and EHV-4 viral DNA. Histological examination confirmed no evidence of EHV lesions in foal tissues and no placentitis in four samples of chorion. There was a mild infiltration of lymphohistiocytic cells in hepatic portal areas and occasionally in the parenchyma. The adrenal gland had an unusual, very slight, diffuse mononuclear cell infiltration in the cortex. Specific lesions were not identified in other tissues, including kidney. Positive leptospiral immunostaining was identified in organisms in the adrenal cortex, kidney and liver; some were a characteristic spiral shape (Fig 1). No organisms were seen in the chorion. The Leptospira PCR on the kidney sample produced a positive Ct value of 30·4, consistent with the presence of Leptospira DNA; the DNA sequence had a 100 per cent identity match with pathogenic Leptospira species recorded in the NCBI database.


Adrenal cortex of a stillborn foal (case 1), showing positive immunostaining for Leptospira. x 400

In case 2, a 15-year-old thoroughbred mare that was vaccinated against EHV aborted at 252 days' gestation after showing no illness or signs of impending abortion. The mare had been on the premises for several years and had had an early pregnancy loss in the previous breeding season. The fetus was rather thin and had pale mucosae, a slightly distended abdomen, perineal meconium staining, a small umbilical hernia and a 5 cm long navel stalk. It showed early autolysis. Blood in the abdomen, a site of hepatic softening and some broken ribs suggested that it had been trodden on by the mare after the abortion. The lungs were firm and there was a reduced amount of coronary fat. Other viscera were macroscopically unremarkable. The chorion was of normal shape. The tip of the non-pregnant horn had torn off and was more autolysed than the rest of the placenta, suggesting that it had been retained for a period. The cervical star was intact and there was a tear across the body. The villous surface had a generally pale, granular appearance: the rostral regions were more congested. There was no overlying exudate or thickening to suggest placentitis. The amnion was slightly opaque but otherwise appeared normal. The length of the umbilical cord was within normal limits but the amniotic portion was rather oedematous and thicker than usual, with several pieces of white fibrinous debris attached to the surface, and a few pale plaques. Bacteriology yielded no growth in aerobic culture of liver or lung swabs. Pseudomonas luteola was isolated from the chorion. PCR analysis of fetal and placental tissues proved negative for EHV-1 and EHV-4 viral DNA, and histological examination confirmed no evidence of EHV-associated lesions. The fetal hepatic sinusoids were dilated and there was a slight to moderate infiltration of portal areas with mainly mononuclear cells. In the lung, neutrophils were seen occasionally within bronchioles, but otherwise the lung tissue was unremarkable. The adrenal gland had a normal appearance except for intense congestion of the medulla. No other specific changes were noted in tissues from the main organs. Histological examination of chorion from five random sites from the horn, body and cervical star all showed a chronic, medium-grade, diffuse, mainly lympho histiocytic infiltrate. The inflammatory cell infiltration extended into many villous cores and the subvillous stroma, and less frequently extended to the allantoic surface. Examination of the macroscopically abnormal area of umbilical cord revealed funisitis (a fibrinous exudate on the surface of the umbilical tunic with an accompanying neutrophil infiltration). Immunohistochemistry revealed large numbers of organisms staining positive for Leptospira on the periphery of the umbilical cord in the region of funisitis (Fig 2) and also associated with the villi and subvillous stroma of the chorion. However, no organisms were seen in the fetal tissues. Leptospiral DNA was not identified in the kidney sample from case 2 that was submitted for Leptospira PCR.


Surface of the amniotic umbilical cord from an equine abortion (case 2), showing a superficial inflammatory exudate containing areas of dark brown particulate immunostaining. x 100. Inset: Higher power enables visualisation of individual positive-staining leptospiral organisms (arrow). x 400

In case 1, the stillborn foal, Leptospira infection was identified both by immunohistochemistry and by PCR analysis. This appeared to be a fetal infection, particularly involving the kidney, liver and adrenal gland, but there was no evidence of placentitis. Case 2 presented differently, in that an unusual diffuse, chronic placentitis and funisitis had led to some growth retardation and then abortion. In this case, immunohistochemistry revealed the presence of organisms associated with the funisitis and placentitis, but no organisms were revealed in the fetal tissues; this correlated with the negative result for PCR on the fetal kidney. Placental tissue was not submitted for leptospiral PCR analysis in case 2. Giant cell hepatitis has been associated with some cases of leptospiral abortion (Wilkie and others 1988) but was not present in either of the present cases. Funisitis has been recognised in association with Leptospira-related abortion in horses (Sebastian and others 2005).

It is believed that these are the first two cases of leptospiral stillbirth/abortion to be identified in England, although, in view of the subtle histological changes in case 1, it is possible that other cases could have gone undiagnosed. The authors suggest that in cases where causes of abortion and stillbirth cannot be determined after routine pathological investigation, the possibility of Leptospira-associated abortion should be considered, particularly if there is any evidence of diffuse placentitis or funisitis. The lack of prominent histological changes in the fetal tissues in cases of leptospiral equine abortion has been noted previously (Donahue and others 1991, Poonacha and others 1993).

The different findings in these two cases indicate that, when investigations of equine pregnancy loss include PCR screening for leptospiral infection, the tissues sampled should include tissues from several internal organs (liver, kidney, adrenal gland) as well as placenta (chorion, cord surface and amnion). Although confirming the presence of leptospires in some lesional sites in these two cases, the methods used gave no indication as to the infecting serovar(s). This information is important for control measures; hence, future studies need to establish the Leptospira serovars that may be involved in equine abortion in England.


View Abstract


Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.