Not Found

Find information on animal health topics, written for the veterinary professional.

Overview of Heartwater


By A. S. Shakespeare, BSc (Eng)(Natal), BVSc (hons) (Pret), MMedVet (Med)(Pret), University of Pretoria

Heartwater is an infectious, noncontagious, tickborne rickettsial disease of ruminants. The disease is seen only in areas infested by ticks of the genus Amblyomma. These include regions of Africa south of the Sahara and the islands of the Comores, Zanzibar, Madagascar, Sao Tomé, Réunion, and Mauritius. Heartwater was introduced to the Caribbean, and it and its vector (A variegatum) are endemic on the islands of Guadeloupe and Antigua. A variegatum, but not the rickettsia, has since spread to several other islands despite attempts at eradication. Possible spread to the mainland threatens the livestock industry of regions from northern South America to Central America and the southern USA. In heartwater endemic areas in southern Africa, it is estimated that mortalities due to the disease are more than double those due to bacillary hemoglobinuria (red water, see Bacillary Hemoglobinuria) and anaplasmosis (see Anaplasmosis) combined. Cattle, sheep, goats, and some antelope species are susceptible to heartwater. In endemic areas, some animals and tortoises may become subclinically infected and act as reservoirs. Indigenous African cattle breeds (Bos indicus), especially those with years of natural selection, appear more resistant to clinical heartwater than B taurus breeds.

Etiology and Transmission:

The causative organism is an obligate intracellular parasite, previously known as Cowdria ruminantium. Molecular evidence led to reclassification of several organisms in the order Rickettsiales, and it is now classified as Ehrlichia ruminantium. Under natural conditions, E ruminantium is transmitted by Amblyomma ticks. These three-host ticks become infected during either the larval or nymphal stages and transmit the infection during one of the subsequent stages (transstadial transmission). The progeny of an infected female tick are most probably not infective (ie, there is no epidemiologically significant transovarial transmission). This and the fact that ticks are indiscriminate feeders probably play a role in the low infection rate in tick populations.

E ruminantium can be propagated experimentally by serial passage, either by inoculating infective blood into, or by feeding infected nymphal or adult stages of a vector tick on, susceptible animals. The organism can also be propagated in tissue culture, most reliably in endothelial cells, but also in primary neutrophil cultures and macrophage cell lines. At room temperature, infective material loses its infectivity within a few hours, but the organism, together with suitable cryoprotectants, may be viably preserved in liquid nitrogen for years.

Immunity to heartwater appears to be chiefly, if not exclusively, cell mediated, because spleen cells from an immune donor inoculated into susceptible recipients protects, whereas serum from an immune donor fails to protect recipients when challenged. There is no, or only partial, cross-protection between different stocks (strains) of E ruminantium. Most of these stocks are infective for, but cannot be serially passaged in, mice; however, a few are pathogenic to mice infected by the IV route.


The pathogenesis of heartwater has not been elucidated; however, the tick probably infects the host via organisms in the saliva or regurgitated gut content while feeding. Replication of the E ruminantium organisms in the tick probably occurs in the intestinal epithelium and is significantly amplified. Once in the host, the organisms may replicate first within the regional lymph nodes with subsequent dissemination via the bloodstream to invade endothelial cells of blood vessels elsewhere in the body. In domestic ruminants, there does seem to be a predilection for endothelial cells of the brain. Organisms can often be found in colonies (commonly but mistakenly referred to as morulas) within the cytoplasm of endothelial cells. Colonies can vary in size, as can the organisms that reside in them. Generally, small-sized organisms are found in larger colonies and vice versa. The smaller organisms are usually referred to as elementary bodies and represent the infective stage, the larger organisms as reticulated bodies and the proliferative stage, and those in between as intermediate bodies.

During the febrile stage, and for a short while thereafter, the blood of infected animals is infective to susceptible animals if subinoculated. Signs and lesions are associated with functional injury to the vascular endothelium, resulting in increased vascular permeability without recognizable histopathologic or even ultrastructural pathology. The concomitant fluid effusion into tissues and body cavities precipitates a fall in arterial pressure and general circulatory failure. The lesions in peracute and acute cases are hydrothorax, hydropericardium, edema and congestion of the lungs and brain, splenomegaly, petechiae and ecchymoses on mucosal and serosal surfaces, and occasionally hemorrhage into the GI tract, particularly the abomasum. The typically straw-colored effusions are high in large-molecular-weight proteins, including fibrinogen; hence, this fluid readily clots on exposure to air. The amount of effusion seen, particularly in body cavities, is not necessarily proportionate to the concentration of parasitic colonies detected in endothelial cells.

Clinical Findings:

The clinical signs are dramatic in the peracute and acute forms. In peracute cases, animals may drop dead within a few hours of developing a fever, sometimes without any apparent clinical signs; others display an exaggerated respiratory distress and/or paroxysmal convulsions. In the acute form, animals often show anorexia and depression along with congested and friable mucous membranes. Respiratory distress slowly develops along with nervous signs such as a hyperaesthesia, a high-stepping stiff gait, exaggerated blinking, and chewing movements. Terminally, prostration with bouts of opisthotonus; “pedaling,” “thrashing,” or stiffening of the limbs; and convulsions are seen. Diarrhea is seen occasionally. In subacute cases, the signs are less marked and CNS involvement is inconsistent.


In clinical cases, heartwater must be differentiated from a wide range of infectious and noninfectious diseases, especially plant poisonings, that manifest with CNS signs. In acute clinical cases in endemic areas, clinical signs alone may suggest the etiology, but demonstration of colonies of organisms in the cytoplasm of capillary endothelial cells is necessary for a definitive diagnosis. Traditionally, this is done with “squash” smears of cerebral or cerebellar gray matter stained with Romanowsky-type stains. Low concentration Giemsa stain developed for 30 min gives the best color differentiation and batch-to-batch consistency. Organisms in autolyzed material lose their stainability, and diagnosis then becomes difficult.

For the “brain squash smear,” a piece of gray matter (~3 × 3 mm) is macerated between two microscope slides; the softened material is then spread like a blood smear with the material pushed rather than pulled along. A slight lifting of the spreader slide about every 5–10 mm creates several thick ridges across the slide, from which capillaries are arranged straight and parallel in the thin sections of the smear for easier examination. The endothelial cells of all the capillaries on a smear should be carefully scrutinized for presence of the dark purple colonies made up of clusters of individual organisms (granules) of E ruminantium. The size of the granules can vary between animals, or smears from the same animal, or even between colonies on the same smear, but is usually uniform within a particular colony.

Using immunoperoxidase staining methods, a definitive diagnosis can be made on any formalin-fixed tissue samples, even from autolyzed carcasses. The contrasting color makes the search for and identification of the rickettsial colonies much quicker, although the substructure of the colonies should be identified before the diagnosis is confirmed. Because of the nature of the test, false-positive reactions may arise with some closely related organisms. On brain squash smears, Chlamydia pecorum can be confused with E ruminantium, but histopathology or the immunoperoxidase technique allow differentiation. Serodiagnosis of animals previously exposed to the disease, ie, recovered from subclinical or clinical infection, still poses problems. Several tests are in use, including several indirect fluorescent antibody and ELISA tests. All serologic tests, including an ELISA that uses recombinant antigen, are plagued by cross-reactions with sera from animals infected with one of several Ehrlichia or Anaplasma organisms (false positive) and the fact that immune cattle on repeated exposure may become seronegative (false negative). DNA probes, available at research institutions, can be used together with PCR technology. A combination of a pCS20 probe and probes to 16S ribosomal RNA of several of the stocks are used routinely to examine samples from animals when permits for movement of animals from endemic to nonendemic areas are required. Real-time PCR has also come into use.

Treatment, Control, and Prevention:

Oxytetracycline at 10 mg/kg/day, IM, or doxycycline at 2 mg/kg/day will usually effect a cure if administered early in the course of heartwater infection. A higher dosage of oxytetracycline (20 mg/kg) is usually required if treatment begins late during the febrile reaction or when clinical signs are evident. In such cases, the first treatment should preferably be given slowly IV. A minimum of three daily doses should be given regardless of temperature; if fever persists, oxytetracycline treatment should continue for a fourth and fifth day. If the fever still does not abate, a potentiated sulfonamide at 15 mg/kg/day, IM, has been successful. The withdrawal times for milk and meat after treatment with doxycycline, short- or long-acting oxytetracycline, and sulfonamides must be observed based on local regulations.

Corticosteroids have been used as supportive therapy (prednisolone 1 mg/kg, IM), although there is debate as to the effectiveness and rationale for their use.

Diazepam may be required to control convulsions.

Affected animals must be kept quiet in a cool area with soft bedding and be totally undisturbed; any stimulation can preempt a convulsive episode and subsequent death.

Vaccination can help with the control of heartwater; however, it is neither easily administered nor monitored and gives variable to no cross-protection to the various E ruminantium stocks. The “infection and treatment method” for immunization is in use in southern Africa, where infected sheep blood containing fully virulent organisms of the Ball 3 stock is used for infection, followed by monitoring of rectal temperature and antibiotic therapy after a fever develops. In certain circumstances, the “controlled” infection is followed by preventive “block treatment” without temperature recording (cattle on day 14 [susceptible B taurus breeds] or day 16 [for the more resistant B indicus breeds], sheep and Angora goats on day 11, and Boer and crossbreed goats on day 12). Young calves (<4 wk old), lambs, and kids (<1 wk old) have an innate age-related resistance to heartwater, so if challenged by natural or induced infections within this time period, most recover spontaneously and develop a reasonable immunity.

Control of tick infestation is a useful preventive measure in some instances but may be difficult and expensive to maintain in others. Excessive reduction of tick numbers, however, interferes with the maintenance of adequate immunity through regular field challenge in endemic areas and may periodically result in heavy losses.

Chemoprophylaxis involves a series of oxytetracycline injections to protect susceptible animals from contracting heartwater when introduced into endemic areas while also allowing them to develop a natural immunity.

Resources In This Article