CASE NOTES

Three outbreaks of Listeria ivanovii abortion in sheep associated with decaying plant matter

Tanya Rajkumar, Katie Boulton (final year veterinary students University of Sydney) and Elizabeth Braddon (Lachlan LHPA DV, Young) and Ian Masters (Hume LHPA DV, Gundagai) and Helen McGregor (Hume LHPA DV, Wagga Wagga) and Michael Hornitzky (Senior Principal Research Scientist, EMAI) and Bruce Watt (Tablelands LHPA DV, Bathurst)

Posted Flock & Herd December 2011

Introduction

Within the Listeria genus, Listeria monocytogenes and Listeria ivanovii are the most important pathogenic organisms in animals, with sheep being more susceptible to infection than cattle (Quinn et al. 2002; Brugere-Picoux 2008). In sheep, infection with L. monocytogenes can cause abortion, encephalitis, meningitis, septicaemia and gastroenteritis (Hughes 1975; Quinn et al. 2002; Sahin & Beytut 2006; Brugere-Picoux 2008). Typically, abortions occur sporadically and other clinical signs of disease are absent in aborting ewes (Hughes 1975; Buxton & Henderson 1999; Mearns 2007). Comparatively, published reports of disease following L. ivanovii infection remain limited. In sheep, this organism has been associated with abortion and the birth of stillborn or weak neonates (Hughes 1975; Sergeant, Love & McInnes 1991; Buxton & Henderson 1999; Chand & Sadana 1999; Quinn et al. 2002; Sahin & Beytut 2006; Mearns 2007).

This paper reviews three abortion outbreaks in sheep in NSW caused by L. ivanovii. These outbreaks are compared and referred back to those in the literature, with the aim of identifying consistent features relating to the duration, timing, location and extent of outbreaks, signalment and clinical presentation of ewes, diagnostic findings, disease management, potential sources of infection and predisposing factors.

CASE 1 - Uranquinty

An outbreak of abortion in a mob of approximately 300 composite ewes in Uranquinty, NSW, was investigated in late July 2011. A total of 30 late-term abortions occurred intermittently over an approximate four week period, from late June to late July, in ewes that were due to commence lambing on the 1^st August. The ewes did not exhibit any other clinical abnormalities.

Four sets of aborted foetuses were submitted for gross pathology, histopathology and culture. On gross examination, several of the placentas were thickened with pale discolouration and flocculent material adhered to their surfaces. Increased serosanguinous peritoneal fluid was reported. Multiple pin point to 3mm focal pale areas were observed throughout the livers of the foetuses, and in the heart of one foetus. On histopathology, multifocal areas of necrosis with a neutrophilic infiltrate and fibrinous material were present within the sections of liver, heart, brain and intercotyledonary placental areas. All submitted samples from the liver, brain, stomach contents and placenta of all foetuses cultured heavy, pure growths of L. ivanovii; the isolates were sensitive to penicillin, amoxicillin, tetracycline, sulfamethoxazole-trimethoprim, erythromycin and neomycin. Abortion due to L. ivanovii infection was confirmed.

The remainder of the mob was treated with long-acting oxytetracycline. The management plan involved prompt daily removal of any further aborted foetuses and placental membranes and running the ewes at a lower stocking rate with the aim of reducing cross-contamination.

The mob of ewes was run on predominantly improved pasture and had been grazing wheat for the duration of the outbreak. The sheep had also had access for one week to weather damaged and mice-infested dumped canola meal dumped in a paddock on the property. Pregnant heifers also had access to this spoiled canola meal for 3 days, but no abortions or stillbirths were reported in this mob.

CASE 2 - Monteagle

On the 29^th July 2011, ultrasound scanning confirmed that 12 Merino ewes from a mob of 750 in the Monteagle district either had aborted or were in the process of aborting. The ewes ranged from 3 to 5 years old, were in good body condition and exhibited no other clinical signs of illness. Abortions continued to affect the mob and an autopsy was performed on two freshly aborted foetuses on the 9^th August 2011. Despite the livers appearing pale compared to normal, no other obvious gross lesions were noted in the specimens. Liver, lung and brain tissue samples were submitted for histopathology, revealing a severe multifocal necrotising hepatitis and mild multifocal suppurative pneumonia. Bacterial colonies were observed in the liver and lung tissue, as well as within the vessels of the brain. A pure growth of L. ivanovii was cultured from the lung and liver samples, confirming a diagnosis of abortion due to L. ivanovii. By the 20^th August, the other three mobs in the flock (1950 in total) had also become affected, with ewes aborting at around 2 to 4 weeks out from lambing.

All mobs were treated with chlortetracycline in the feed for 5 to 7 days, with the aim of preventing further cases. A total of 4.5 kg of antibiotic was added to the feed per day equating to 600mg of chlortetracycline per feeding dose. No more abortions were reported in the flock after day 4 of treatment. An estimated 70 ewes had aborted during the outbreak which lasted from 27^th July to 25^th August.

The suspected source of infection was decaying vegetable matter and litter left over from the summer period. During late summer, Monteagle had encountered significant rain and flooding, while June had also been quite wet. The ewes did not have any other access to poor silage or spoilt feed.

CASE 3 - Cootamundra

An outbreak of abortion in a mob of 260 maiden Merino ewes in Cootamundra NSW was investigated in early August 2011. A total of 10 to 15 lambs within a 4 day period (approximately from the 5^th to 8^th August) were aborted 2 to 4 weeks prior to their due date. The abortions stopped after this time. No other clinical abnormalities were observed in the ewes and no problems were noted in older ewes on similar feed.

Results of one on-farm foetal necropsy performed on the 8^th August revealed an enlarged liver with numerous pin point white lesions on the surface. The foetal stomach could not be examined as it had been attacked by crows. The liver of this foetus (Lamb 1) and a second entire foetus enclosed within foetal membranes (Lamb 2) were submitted for histopathology and bacterial culture. On necropsy of Lamb 2, scattered 1mm diameter poorly defined white areas were present on the surface of the liver. No other gross lesions were identified except blood-stained subcutaneous fluid consistent with death in-utero. As the stomach was empty, stomach contents could not be sampled. Histopathology revealed numerous scattered colonies of small bacilli throughout the autolysed liver tissue. The liver of Lamb 1 was better preserved and a morphological diagnosis of acute, multifocal, necrotising, suppurative hepatitis with intra-lesional bacteria was made.

Profuse pure growths of L. ivanovii were cultured from the livers of Lamb 1 (isolate definitively identified) and Lamb 2 (isolate identity presumed). A diagnosis of abortion due to L. ivanovii was confirmed.

The ewes had recently been moved to a rank phalaris dominant pasture, with new green shoots at the base. This pasture had been slashed earlier in the year to remove summer growth. The ewes had not been fed silage. An accumulation of spoiled pasture combined with wet, cloudy weather was suspected to have provided optimal conditions for Listeria growth.

Discussion

To date there have been few reports of L. ivanovii as a cause of abortion amongst livestock in Australia, although it has been associated with isolated cases of neonatal death and sporadic abortion amongst both sheep and cattle (Sergeant, Love & McInnes 1991; Gill et al. 1997).

The three cases reported here occurred in the Uranquity, Monteagle and Cootamundra districts in the south west slopes of NSW. The abortion outbreaks all took place between June and August of 2011, with the ewes exhibiting no other clinical signs of illness. This concurs with previous reports of L. ivanovii abortion, as infected ewes typically appear otherwise healthy and in good condition (Chand & Sadana 1999; Sahin & Beytut 2006; Mearns 2007). In the cases reported here, abortions occurred 2 to 4 weeks prior to lambing. This is consistent with previous reports of abortions usually occurring within the last trimester of gestation (Buxton & Henderson 1999). Affected ewes ranged from three to five years of age.

The duration of outbreaks varied. The outbreak in Cootamundra (Case 3) only lasted 4 days while those in Uranquinty (Case 1) and Monteagle (Case 2) occurred over 4 weeks. Outbreaks reported by Sergeant, Love and McInnes (1991) and Chand and Sadana (1999) also occurred over 4 and 6 weeks respectively.

The percentage of the flock affected was similar in both Case 2 and Case 3, with approximately 3.5% and 4 to 6% of ewes aborting respectively. However, in Case 1, an abortion rate of 10% was reported. Higher incidence rates have also been described overseas, with Chand and Sadana (1999) reporting 23% losses in a flock in northern India. Interestingly, in Case 1, no abortions were noted amongst the herd of pregnant heifers that also had access to the spoiled canola meal for 3 days. This finding further supports earlier claims that sheep appear more susceptible to listeriosis than cattle (Brugere-Picoux 2008).

Gross abnormalities of the liver were reported in all three cases. In two cases, multifocal pale to white areas, approximately 1mm to 3mm in size and located on the surface (Case 3) and throughout (Case 1) the examined foetal livers, were noted. Similar lesions were distributed on the heart of one foetus. In case 2 a more generalised pale discolouration of the liver was noted, while an enlarged liver was also observed in one case.

These findings are consistent with other reports of ovine abortions, following L. ivanovii infection. Sahin and Beytut (2006) described yellowish to white necrotic foci distributed on the liver surfaces, predominantly on the right lobes in five aborted ovine foetuses. Similarly, multiple creamy spots were identified scattered throughout the liver parenchyma of four aborted ovine foetuses from a case in NSW (Sergeant, Love & McInnes 1991). Chand and Sadana (1999) described yellowish to white necrotic foci on the surface of the right liver lobes of seven foetuses, and throughout the parenchyma of one foetus. The liver lesions of aborted foetuses in these reports are of a similar colour and size (ranging from 0.5mm to 3mm) as the lesions reported in Cases 1 and 3 (Sergeant, Love & McInnes 1991, Chand & Sadana 1999 and Sahin & Beytut 2006).

Thickening of the placental membranes and the presence of adherent flocculent material were observed in Case 1. Previous reports have generally not commented on the gross appearance of placentas (Sergeant, Love & McInnes 1991; Chand & Sadana 1999; Sahin & Beytut 2006). Other findings of the three cases in this paper, including a marked increase in serosanguinous peritoneal fluid and the presence of blood-stained subcutaneous fluid, have been described in the literature associated with ovine and bovine abortions following L. ivanovii infection (Sergeant, Love & McInnes 1991; Gill et al. 1997; Sahin & Beytut 2006). However, as in Case 2, lesions associated with listeriosis aborted foetuses or stillbirths may be absent or minor on gross examination (Hughes 1975; Sergeant, Love & McInnes 1991; Mearns 2007). Other gross abnormalities reported in aborted foetuses, including erosions of the abomasal mucosa, mesenteric lymphadenopathy and congestion of organs (spleen, kidneys and adrenal glands) were not observed in the three cases outlined in this paper (Chand & Sadana 1999; Mearns 2007).

Ovine listeriosis is known to result in multifocal areas of necrosis associated with the liver, spleen and myocardium (Brugere-Picoux 2008). The three cases reported in this paper demonstrated consistent changes on histopathology of the liver, with a multifocal necrotising hepatitis described. A suppurative, fibrinous component may be present, as evidenced by these cases. Similar lesions were identified in the heart, brain and intercotyledonary placental areas in Case 1, and in the lungs in Case 2. These findings associated with the liver are similar to changes reported by Sahin and Beytut (2006) and Sergeant, Love and McInnes (1991), although Gill et al. (1997) described mild to moderate lympho-histiocytic portal inflammation without necrotic lesions. Pathology of the lungs, heart and placenta has been observed histologically in ovine and bovine cases of L. ivanovii abortion (Gill et al. 1997; Sahin & Beytut 2006). Sahin and Beytut (2006) reported lung lesions similar to those described in Case 2 including a multifocal necrotising haemorrhagic pneumonia. Bacterial colonies were observed in two cases in this paper, distributed in the lung and liver parenchyma, and within the vessels of the brain. Bacterial aggregations in liver and lung samples, with some engulfment by macrophages, have been reported previously (Sergeant, Love & McInnes 1991; Sahin & Beytut 2006).

All three cases described in this paper confirmed a diagnosis of L. ivanovii infection through isolation of bacterial colonies exhibiting features consistent with this organism; small (3 to 4mm after 4 days), circular, smooth, whitish-grey - haemolytic colonies on blood agar (Chand and Sadana 1999). Previous reports have cultured L. ivanovii from samples of ovine and bovine liver, lungs and stomach contents (Sergeant, Love & McInnes 1991; Chand & Sadana 1999; Gill et al. 1997; Sahin & Beytut 2006). In addition to isolation from these tissues, this report also notes successful recovery of the organism from other affected organs, namely the brain and placenta.

Two of the properties in this report treated the affected mobs with tetracycline (oxytetracycline and chlortetracycline), with the aim of preventing further cases. It is unknown how long abortions continued after commencement of treatment at the Uranquinty property, but no cases were reported at Monteagle after day 4 of treatment. Considering isolate sensitivity reported in Case 1, penicillin, amoxicillin, sulfamethoxazole-trimethoprim, erythromycin and neomycin may be alternate treatment options. Treatment for the encephalitic form of listeriosis has been described, including therapeutic use of ampicillin, gentamicin, procaine penicillin and tetracycline (Brugere-Picoux 2008). However, conflicting data regarding the effectiveness of treatment of in-contact ewes during abortion outbreaks exists (Buxton & Henderson 1999; Mearns 2007). Culling of aborting ewes may be another consideration, as these can become chronic carriers (Brugere-Picoux 2008). The recommendation in Case 1 for prompt removal of aborted foetuses and placental membranes from pastures is justified as aborted foetuses are a potential source of infection for the remainder of the flock (Brugere-Picoux 2008).

L. ivanovii, like L. monocytogenes, is ubiquitous although a number of conditions have been reported to favour its growth (Hughes 1975). These include cold, wet weather and feeding of spoiled plant matter, particularly poorly fermented silage (Hughes 1975; Sergeant, Love & McInnes 1991; Gill et al. 1997; Sahin & Beytut 2006; Mearns 2007; Brugere-Picoux 2008). These cases are similar in that they occurred following wet weather and the affected mobs had access to spoiled plant matter. In Cootamundra, the likely source was rank phalaris pasture, at Uranquinty weather-damaged canola meal while decaying vegetable litter was suspected to be the contaminant in Monteagle. The source of contamination could be verified by culturing feed samples, although this was not done in these cases. Sergeant, Love and McInnes (1991) report successful isolation of L. ivanovii from rotting lucerne hay considered to be the infection source in an abortion outbreak at Manilla, NSW.

No silage was fed in any of the three cases discussed in this paper, demonstrating that decomposing plant litter and feed also provide suitable sites for the survival and growth of L. ivanovii. Several other predisposing factors for the disease have also been suggested. Poor nutritional status, overcrowding, unhygienic conditions, contaminated water sources, confinement in sheds and the stress associated with advanced pregnancy have all been recognised as potential risk factors (Sergeant, Love & McInnes 1991; Chand & Sadana 1999; Santagada 2004; Sahin & Beytut 2006; Brugere-Picoux 2008).

References