Ross River virus infection

INTRODUCTION — Ross River virus (RRV) is a group A arbovirus transmitted by mosquitoes that causes a disease manifested by polyarthritis and rash. The illness was first described in northern Australia in 1928 and subsequently has been observed widely through Australia and many islands of the western South Pacific. It is the most prevalent vector-borne disease in Australia. The virus can be transmitted by many mosquito species and survives in mosquito eggs in arid environments. Therefore, it has the capacity to spread to other geographic areas.

Other viruses that cause arthritis are discussed separately. (See "Viral arthritis: Causes and approach to evaluation and management".)

EPIDEMIOLOGY — RRV infection has been described in mainland Australia [1], Papua New Guinea [2], the Solomon Islands, the islands of eastern Indonesia, the western South Pacific [3,4], Fiji [5], and American Samoa [6]. As many as 500,000 people were infected in Fiji when the disease first appeared in 1979 [5]. In 2013, seroprevalence in Fiji residents was 46 percent; in 2015, the seroprevalence was 57 percent, demonstrating ongoing endemic circulation [7]. Outbreaks have been described in American Samoa, New Caledonia, French Polynesia, the Cook Islands, and most states of Australia [8,9]. Over 30 percent of blood donors in French Polynesia are seropositive [10]. Genomic characterization of the first isolate from Papua New Guinea indicates that RRV had evolved over approximately 40 years; the most recent common ancestor coincides with the beginning of the 1979 Pacific Island epidemic [11]. A serological survey in Papua New Guinea in 2019 of military personnel indicated 93 percent positivity [12].

In Australia during the 2014 to 2015 season, 84 percent of reported mosquito-transmitted infections were due to RRV and Barmah Forest Virus infection [13]. There is an increasing incidence in Australia and the Pacific Islands; 9544 cases were reported in the largest Australian outbreak in 2015, with southeast Queensland most affected [14]. It is anticipated that RRV disease will increase with extension into temperate Australia and longer active periods of infection [15]. While the highest rates of RRV infection have been reported from the Northern Territory [16], endemic foci also occur in more temperate eastern Australia. Southeast Queensland with its larger population bears a particularly high public health and economic burden [17]. Infection has been observed in the outskirts of Sydney since 1997 [18]. Between 2006 and 2009, increasing modifications were noted from the Perth metropolitan area in off-seasons (June to September) [19]. Epidemics have increased in frequency, size, and range throughout Australia [20]. The geographic range is expanding in Queensland [21] and South Australia [22]. Transmission is most concentrated in areas where urban and periurban environments intersect [17].

RRV infection occurs most frequently in tropical coastal regions with salt marsh habitats suitable for the principal mosquito vector species [23]. The infection is most common in spring, after summer rains, or following inundation of salt marshes by rain or tides. The virus persists in arid areas of inland Australia within desiccation-resistant mosquito eggs. Outbreaks occur after rain, when mosquitoes hatch [24].

Climate influences disease activity. Models predicting disease activity differ among regions; the relationships between climate, social, and environmental factors are complex. In the River Murray Valley in Southern Australia, important factors include river height, rainfall, and mosquito abundance [25]. Flooding is significantly associated with transmission [26]. In a Tasmanian study, a combination of high spring tides and excessive summer rainfall produced a salt marsh habitat with a high density of mosquitoes [27]. Isolation of RRV from these mosquitoes coincided with the onset of the first human cases of RRV infection. In Western Australia, climate data were moderately sensitive in predicting epidemics (64 percent); the addition of mosquito surveillance data improved sensitivity of an early warning model for epidemics (90 percent) [28]. Transmission peaks at moderate temperatures (26°C) and declines within thermal limits (17°C to 31.5°C); this explains in part endemic year-round transmission in the tropics, but seasonal transmission in temperate zones [29]. Accurate models for predicting infection rates would be useful for strategies to control mosquito vectors [30-32]; in one review of RRV predictive models, models included climate and weather, mosquito abundance, rainfall, temperature, and tide height [33].

Dryland salinity, a substantial and expanding problem in the wheat belt of Southwestern Australia, is associated with an abundance of Aedes camptorhynchus and increased potential for zoonotic transmission [34], although increased human seroprevalence has not yet been demonstrated [35]. Important factors for infection in the Darwin region of tropical Northern Australia include rainfall, minimum temperature, and abundance of three mosquito species [36]. The relative importance of these environmental factors differs according to region, although rainfall is consistently associated with increased disease activity [37,38].

Weather extremes, such as intense summertime rain events, are expected to increase with climate change and may lead to significant increased arboviral disease [39].

Travelers to affected areas of Australia have presented with illness on returning home [40].

RRV is also transmissible by blood transfusion [41]. However, studies estimating the risk of transmission by blood transfusion are reassuring. Of 7500 blood donations from selected collection centers in Australia during peak transmission season, none had detectable RRV RNA [42,43]. Existing risk management strategies include donation restriction and recalls.

New techniques of virus tracking through molecular testing of mosquito trap-derived specimens offer potential for fast detection and characterization of RRV isolates, resulting in improved epidemiological information [44].

Vectors and hosts — RRV has multiple vectors and multiple hosts. The mosquito species serving as vectors vary by region; RRV has been identified in 40 different mosquito species [30]. In coastal Australia, Aedes vigilax (salt marsh mosquito) and Ae. camptorhynchus are major vectors; both of these species feed during the day. In inland areas except Tasmania, Culex annulirostris predominates; this species feeds at night [45]. Other species capable of transmitting RRV include Aedes tremulus, Aedes polynesiensis, Aedes aegypti, Culex gelidus, and others [46-48]. A. aegypti has the potential to extend the distribution of this infection. Nonhuman vertebrates are critical for the maintenance and spillover of RRV into mosquito populations. In a Malaysian RRV outbreak, Ae. aegypti and Ae. albopictus were identified as vectors, with Ae. albopictus showing higher vector competence [49].

Native macropods (eg, kangaroos and wallabies) are the principal vertebrate hosts [50]. In southwest Western Australia, over 40 percent of western grey kangaroos have neutralizing antibodies to RRV and are believed to have a role in maintenance and transmission of RRV [51].

Infection in koalas in southeast Queensland has been studied [52], with infection rates exceeding 80 percent. The koalas are now confined in this region to the edges of permanent wetlands unsuitable for urban development, with large numbers of mosquito vectors. Previously, it was felt marsupials were the most important viral reservoir, but high rates of infection in Fiji where there are no marsupials indicate that they are not essential for endemic infection [53]. Neutralizing antibodies were found in 28 to 100 percent of Fijian sera from horses, cows, pigs, dogs, cats, mice, and rats.

Infection also occurs among other wild animals and livestock. Possums and horses are potential reservoirs in urban areas [54]. RRV is endemic in horses in Queensland, Australia [55]. Apart from humans, horses are the only other known host showing clinical manifestations, including fever, malaise, polyarthralgia, joint swelling, and reduced athletic performance [56]. Their large body mass and high levels of viremia suggest they may be an important reservoir of the virus. In a field survey of mosquito endemicity in Brisbane, suburbs with higher RRV notification rates had a larger vertebrate biomass (mainly horses) and higher mosquito abundance [57]. The potential of horses as sentinels for arbovirus monitoring has also been suggested [58].

In addition, vertical transmission occurs in mosquitoes, and human-mosquito-human transmission is believed to occur. Studies of RRV in northwest Victoria suggest that the density of short-lived, highly fecund hosts, such as mice, may also influence rural disease transmission [59]. Humans and possums appeared to contribute most to epidemic transmission, followed by macropods [60].

PATHOGENESIS — RRV, an Alphavirus and member of the Togaviridae family, is an enveloped single-strand RNA virus. Distinct genetic types have been found in eastern and western Australia, with a mix in northern Australia. The membrane bilayer, which surrounds the nucleocapsid, is penetrated by spikes, each of which consists of glycoproteins E1 and E2 [61]. E2 is thought to be involved in recognition of the cellular receptor in the host [62]. Envelope glycoprotein glycosylation affected disease outcome in a mouse model [63]. Relapse or persistence of symptoms in many patients may be related to persistent replication within synovial macrophages despite neutralizing antibodies and antiviral cytokine responses [64]. This process may be initiated by an initial immunoglobulin (Ig)G anti-RRV antibody response, which enhances infection in macrophages and monocytes [65]. Chondrocytes may play an important role in pathogenesis, as suggested by recent studies on the pathways and responses elicited by RRV in these infected cells and neighboring infected cells [66].

Host genetics play a major role in influencing susceptibility to infection and severity of clinical disease [67]. RRV has been suggested as a model for study of myalgic encephalomyelitis/chronic fatigue syndrome. Studies have identified immune evasion strategies which may be relevant in viral pathogenesis, including selective proinflammatory gene suppression [68]. Functional polymorphisms in immune-response genes contribute to susceptibility and outcome [69]. Dominant inheritance has been demonstrated in some studies [70]. Polymorphisms in the interferon (IFN)-gamma and interleukin (IL)-10 loci influence cytokine production and the severity and duration of illness after infection with RRV. These risk-associated cytokine genotypes may act in combination to potentiate their impact on illness severity. IL-10 is a potent inhibitor of IL-1-beta and TNF-alpha production by macrophages, and these cytokines may have synergistic activities in induction of inflammation in the acute sickness response. In vitro models suggest that cellular antiviral response can direct selective pressure for viral sequence evolution that impacts on viral fitness and sensitivity to alpha/beta interferon [71]. Immune response, particularly innate immune response, is a major predictor of clinical outcome [67,72]. (See "Clinical features and diagnosis of myalgic encephalomyelitis/chronic fatigue syndrome" and "Treatment of myalgic encephalomyelitis/chronic fatigue syndrome".)

The arthritis of RRV infection is believed to be mediated by local cell-mediated immune responses. In the majority of patients, inflammation is present in the absence of detectable virus, although RRV DNA has been identified in the inflamed synovium of some patients more than a month after the onset of symptoms [73].

The rash of RRV infection may be a manifestation of a cell-mediated immune response to viral antigen in the basal epidermal and eccrine duct epithelial cells [74]. Circulating immune complexes have not been found in patients with persistent symptoms [75]. In a mouse model, macrophages have been shown to be the primary mediators of muscle damage [76].

In patients with prolonged arthralgia and a long convalescent course, increased proinflammatory cytokine levels have been demonstrated [77].

CLINICAL MANIFESTATIONS — RRV infection has been observed in individuals between 4 and 85 years of age; the disease occurs slightly more commonly in women [78]. The incubation period for acute presentations is usually 3 to 9 days but may be as long as 21 days [23]. Subclinical infection is common [79].

Symptomatic RRV infection presents in one of the three ways [78]:

●Acute febrile illness with arthritis and rash

●Acute fever, rash, or arthritis alone

●Chronic polyarthritis or polyarthralgia

There are no laboratory findings specific for RRV infection. The leukocyte count is usually normal; atypical lymphocytes or a mild neutrophilia may be found in some patients. The serum C-reactive protein concentration is usually normal, but modest elevations in the erythrocyte sedimentation rate may be observed.

Arthritis — The most dramatic clinical features of RRV infection are marked arthralgia and myalgia [38]. Joint pains are present in more than 95 percent of patients. True arthritis occurs in 40 percent of patients (picture 1). The joints of the extremities are most frequently involved, especially the wrists, knees, ankles, and the metacarpophalangeal and interphalangeal joints of the fingers (picture 2). Joint involvement is usually symmetric, but the severity of inflammation may be asymmetric. The intensity of the pain and inflammation occasionally may resemble that of gout. Effusions are often present in joints and tendon sheaths, and Achilles tendinitis and plantar fasciitis may occur. Back pain is also common; no association between human leukocyte antigen (HLA)-B27 and another HLA allele has been detected [79].

Synovial fluid analysis early in the disease may suggest viral arthritis [78]. The fluid is viscous with entirely or predominantly mononuclear cells; neutrophils may comprise up to 12 percent of cells.

Rash and fever — Rash and fever are observed in approximately one-half of patients with RRV infection. Rash and arthritis usually develop within two days of one another but may be separated by up to two weeks. The rash may occur before or after the arthritis. It is typically short lived (1 to 10 days) but may persist for several months [78]. The palms, soles of the feet, face, and mucous membranes of the mouth may be involved. The rash usually consists of 1 to 5 mm maculopapular lesions on the limbs and trunk; it is often sparse (picture 3), although in some cases it is dense and can include purpuric areas or small vesicles (figure 1) [80].

Fever, occasionally reaching 39ºC, is present early in RRV infection in about 20 percent of patients and usually resolves within three days. Chills and delirium can occur; rigors are unusual.

Other clinical features — Fatigue is the most frequent constitutional symptom, occurring in 90 percent of patients. It may be prominent, even in the absence of fever, and may persist after other symptoms resolve [81]. Anorexia and headache (50 percent patients) are common. Other symptoms and signs include diarrhea, neck stiffness, photophobia, splenomegaly, aseptic meningitis [82], hematuria [83], glomerulonephritis [2], and superficial lymphadenopathy (usually mild; occasionally tender). Several patients with encephalitis have been described [84-86]; one case of fatal encephalitis has been attributed to RRV infection [84]. RRV has not been associated with congenital abnormalities, although transplacental viral transmission was described [87].

Clinical course — The average duration of symptoms differs considerably in different reports. Some arthralgias persist for more than three months in two-thirds of patients, but it usually steadily diminishes in severity over this time [50,88]. In one study, for example, nonsteroidal anti-inflammatory drug use decreased from 63 percent at one month to 15 percent at three months. Early studies suggested patients can have persistent arthralgia, myalgia, and fatigue for more than a year after infection; more recent evidence suggests that symptoms usually resolve within three to six months [23,89,90]. Many patients with longer duration of symptoms had other underlying comorbidities such as rheumatic disease or depression [91]. Degenerative joint changes do not occur, and relapses are infrequent.

In a prospective cohort study of patients with RRV, Epstein Barr virus, or Q fever infection, 12 percent of patients had prolonged illness with disabling fatigue, musculoskeletal pains, neurocognitive difficulties, and mood disturbance at six months following initial infection [92]. The outcome was predicted by severity of acute illness rather than the underlying microbiologic diagnosis [92].

DIAGNOSIS — RRV infection should be suspected in the setting of acute polyarthritis and rash with a history of travel or residence in endemic or epidemic areas.

Differential diagnosis — The differential diagnosis includes rubella, infectious mononucleosis, parvovirus B19, immunoglobulin A vasculitis (Henoch-Schönlein purpura), serum sickness, a drug reaction, erythema multiforme, and systemic lupus erythematosus. It also includes other arboviral infections, particularly Barmah Forest virus (BFV) infection and chikungunya fever [12]. (See "Chikungunya fever: Epidemiology, clinical manifestations, and diagnosis".)

BFV is an alphavirus transmitted by mosquitoes that causes polyarthritis. BFV causes approximately 10 percent of epidemic polyarthritis in northeastern Australia. There are approximately 2400 notifications annually in Australia, most in the state of Queensland but also large numbers in Western Australia [93]. It was first recognized as a cause of human infection in 1988 and has been increasingly reported in Australia [94]. Epidemics have been recognized in the Northern Territory (1992), Western Australia (1992 to 1993) [94,95], coastal New South Wales (1995), Victoria (2002) [96], and Queensland (2002 to 2003) [16].

Transmission of BFV depends on several environmental factors [97]. Peak transmission occurs in summer and autumn [16]. Endemic activity is believed to be restricted to northern Australia. As for RRV, human land use change is an important driver of disease outbreaks and modifying endemic transmission. The first mosquito isolates of this virus were from Culex annulirostris, but this mosquito species has been shown to be an inefficient vector. Other vector mosquitoes of importance include Ae. vigilax (including an epidemic in 1995), Ae. camptorhynchus (epidemic in 1993), Ochlerotatus camptorhynchus (2002), and Ae. notoscriptus (estimated to have a transmission rate of 45 percent) [30]. The biting midge (Culicoides) is a vector in Western Australia.

Serologic investigation of native animals has demonstrated infection especially in Eastern Grey kangaroos and cattle, but low rates in brushtail possums, koalas, quokka, domestic cats and dogs, and horses.

The clinical manifestations of BFV are similar to those of RRV although, in general, BFV tends to cause milder disease [23]. Rash is more common with BFV infection, while arthritis is more prominent in RRV infection, although these diseases cannot be reliably distinguished based on symptoms alone [89,98]. In one series of 84 patients with BFV, the most common symptoms and signs were lethargy (89 percent), joint pain (82 percent), and rash (68 percent) [98]. Dermatologic manifestations include urticated erythema and widespread vesiculopapular, macular, or purpuric eruptions [96].

Diagnosis of BFV is confirmed by a virus-specific IgM response [99].

Laboratory evaluation — The diagnosis of RRV infection is generally established by serology; the most common serologic methods are the hemagglutination inhibition (HI) antibody test and the enzyme-linked immunosorbent assay (ELISA) test. A fourfold rise in HI titer is diagnostic; levels of ≥1:1280 indicate recent infection. RRV-specific IgM persists for months after acute infection. Thus, acute infection should be confirmed by rising IgM titers if patients live in endemic areas [23]. Testing for a specific IgA response can demonstrate acute infection in patients with persistent high levels of anti-IgM and IgG antibodies, although this test is rarely necessary [100].

Viral culture and polymerase chain reaction (PCR) are less widely used diagnostic tools. RRV can be isolated from blood drawn early in the course of infection [23,78]. RRV has not been cultured from joint fluid, but viral antigen has been detected by immunofluorescence in joint fluid specimens collected early in the disease. A nested PCR is being developed for RRV infection, but it remains a research tool currently [101].

TREATMENT — Treatment of RRV consists of supportive care; no specific antiviral therapy is available. Treatment of the arthralgias and myalgias with analgesics and nonsteroidal anti-inflammatory drugs may be helpful. Some individuals benefit from swimming, hydrotherapy, physiotherapy, or massage; others gain relief only with rest.

PREVENTION — The most important preventive measure is avoidance of mosquito bites [45]. In one case-controlled study, mosquito coils, repellants, and citronella candles were shown to decrease the risk of infection at least twofold; light-colored clothing decreased risk threefold [102]. In contrast, camping increased the risk of RRV eightfold. Screens should be fitted to windows and doors in high-risk areas. Drainage of mosquito breeding areas in endemic areas may also be helpful over the long term to control the infection [8]. Mosquito control programs affect RRV disease rates [103]. Preemptive surveillance based on knowledge of disease habitats has been associated with lower RRV disease rates. (See "Prevention of arthropod and insect bites: Repellents and other measures".)

RRV is transmissible by transfusion of blood products, and asymptomatic viremia has been documented [104,105]. There is no proposal to test donors in affected areas at present, but they are requested to report post-donation illness [106].

No vaccine is available; experimental vaccines are under study [107-109]. An inactivated whole virus vaccine that has proven effective in mice [110] has been shown to be well tolerated and immunogenic in adults in a phase III trial [110,111].

SUMMARY

●Epidemiology and transmission – Ross River virus (RRV) is transmitted by mosquitoes and causes an illness characterized by fever, polyarthritis, and rash. The illness was first described in northern Australia and subsequently has been observed widely throughout Australia and many islands of the western South Pacific. It is common in the tropical coastal regions of northern Australia. RRV can be transmitted by several mosquito species, including Aedes vigilax, Aedes camptorhynchus, and Culex annulirostris. Kangaroos and wallabies are the principal vertebrate hosts. Horses may be important reservoirs in some settings. (See 'Introduction' above.)

●Clinical manifestations – Symptomatic RRV infection typically presents as acute illness with fever, arthritis, and rash. These symptoms and signs frequently occur together but may occur in isolation. Arthritis in the wrists, knees, ankles, and finger joints is common and frequently symmetric. The rash is usually maculopapular and involves the limbs and trunk. Arthralgia may persist for up to three months. Fatigue is the most frequent constitutional symptom. (See 'Clinical manifestations' above.)

●Diagnosis – The diagnosis of RRV infection is generally established by serology; the most common serologic methods are the hemagglutination inhibition (HI) antibody test and enzyme-linked immunosorbent assay (ELISA) test. A fourfold rise in HI titer is diagnostic; levels of ≥1:1280 indicate recent infection. RRV-specific immunoglobulin (Ig)M persists for months after acute infection, and acute infection can be confirmed by rising IgM titers. (See 'Diagnosis' above.)

●Treatment – Treatment consists of supportive care, analgesics, and nonsteroidal anti-inflammatory drugs. (See 'Treatment' above.)

●Prevention – The most important preventive measure is avoidance of mosquito bites. Helpful interventions include use of window and door screens, mosquito repellants, and light-colored clothing. Drainage of mosquito breeding areas in endemic areas may be helpful to control infection. (See 'Prevention' above.)