114 Q Fever

The “Q” in Q fever stands for “query,” the designa­tion applied by E. H. Derrick to an acute illness with fever and severe headache of unknown cause occur­ring in abattoir workers and dairy farmers in Queensland, Australia, in 1935.

Q fever is caused by infection with Coxiella burnetii, previously known as Rickettsia burη,etii, and is the sole member of the genus Coxiella, family Rickettsiaceae. It was initially confused with vi­ruses, but though C. burnetii is an obligate in­tracellular parasite, it has a true bacterial cell wall.

Q fever is a zoonosis of worldwide distribution, and many species of animals, birds, ticks, and other biting insects are natural hosts. In animals, naturally ac- ^, quired infection appears to be asymptomatic so that Q fever is not of any economic significance to farmers. Transmission to humans occurs via inhalation of con­taminated dust while infected animals, carcasses, or animal products are being handled; via laboratory accidents; and sometimes via tick bite and the con­sumption of unpasteurized milk. Asymptomatic in­fection is common. Illness may take two forms. Acute Q fever is usually a self-limiting febrile flulike illness or atypical pneumonia lasting up to 4 weeks. Un­treated, the fatality rate is less than 1 percent. Chronic Q fever may develop months to years later, presenting as endocarditis and/or hepatitis. Endo­carditis usually occurs in those with preexisting heart valve disease; untreated, it is usually fatal.

Distribution and Incidence

The distribution of acute Q fever is worldwide, but chronic Q fever appears to be more limited, most cases being reported from the United Kingdom, Aus­tralia, and Ireland.

It is certain that Q fever is very considerably un­derdiagnosed so that accurate national incidence fig­ures are not available. In the United States by 1960, Q fever was recognized to be endemic in dairy herds throughout the country. Reports of human cases were uncommon but increased steadily from 1948 to 1977. Over this period, 1,164 cases were reported to the Centers for Disease Control. Although most states reported cases, fully 67 percent of them were from California, where there is a very high prevalence of infection in cattle and consequently a high proportion of cattle excreting organisms in milk. Reported infec­tion in humans, however, is relatively rare, with only 148 being reported from 1970 to 1978. Seropreva­lence surveys of cattle in the United Kingdom sug­gest that about 3 to 5 percent of dairy cattle and 2 to 6 percent of sheep have been infected. About 100 to 200 cases are reported by laboratories each year, and 10 percent of these are chronic infections.

Epidemiology and Etiology

Coxiellae differ from other Hckettsiae in not needing an arthropod vector for transmission. Two cycles of natural infection have been recognized. In the wild­life cycle, transmission between wild animals and birds occurs from tick bites, from inhalation of dust, and, possibly in carnivores, from ingestion of infected placentas and meat.

Coxiellae are very resistant to environmental con­ditions, surviving for months or years in dust or animal litter. Experiments have shown that these organisms can survive in 1 percent formalin for 24 hours, on wool for 7 to 9 months at 20^oC, in dried blood for at least 6 months at room temperature, and in tick feces for at least 18 months at room tempera­ture; they can also survive temperatures of 63^oC for up to 30 minutes. The latter observation is impor­tant since it places coxiellae at the border of resis­tance to high-temperature, brief-interval milk pas­teurization. However, there is still dispute about whether symptomatic Q fever can be acquired from drinking contaminated milk, although the preva­lence of antibodies is raised in raw milk drinkers. Person-to-person spread has been documented but is very unusual. Laboratory accidents are a serious hειzard when working with live organisms, and for this reason routine diagnostic laboratories seldom attempt isolation.

The vast number of organisms excreted by in­fected animals, the dissemination of the organisms in dust by wind, the hardy nature of the organism, and the low infective dose for humans (said to be one organism) explain a characteristic feature of Q fever-the occurrence of localized, explosive out­breaks often without an obvious source of infection.

Immunology

The serologic diagnosis of acute, as opposed to chronic, Q fever was considerably helped by the dis­covery of phase variation in the organism. Coxiellae isolated directly from animals or ticks are in phase 1; in this phase, specific antibodies are induced that are detectable by complement fixation and other serologic tests. When coxiellae are passed serially through chick embryo yolk sacs, phase 2 organisms become predominant. Passage of phase 2 organisms through animals causes the predominant phase to revert to phase 1. It has been found that phase 1 antigen reacts with late convalescent serums from inoculated guinea pigs; and phase 2, with early con­valescent serums in guinea pigs.

Within 2 weeks of acute infection in humans, all or most patients will develop IgM, IgG, and IgA antibodies against phase 2 and IgM antibodies against phase 1, detectable by indirect immunofluo­rescence tests, as well as complement-fixing anti­bodies to phase 2. The IgM antibody titer declines rapidly after 2 to 3 months. Phase 1 IgG, IgA, and complement-fixing antibodies may be detectable af­ter about 4 weeks, but do not reach high titers unless chronic infection ensues. In chronic Q fever, there are usually high titers of IgG antibodies detectable by immunofluorescence and complement fixation to both phase 1 and 2. However, IgM antibody is either absent or present in very low titers, and there are high titers of IgG to phase 1.

The presence of IgG antibodies is associated with immunity, but cell-mediated immunity is known to be important as well. The phase 1 antigen used in vaccines gives far better protection than phase 2 vaccines. The intradermal skin test using diluted vaccine may be employed to identify patients who are already immune and who are more likely to have hypersensitivity reactions to the vaccine.

Clinical Manifestations and Pathology

Subclinical infection is common. Only 191 of 415 newly infected persons were symptomatic in an out­break in Switzerland in 1983, caused when infected sheep were driven from mountain pastures through several villages on the way to market. The clinical features of acute Q fever are shared by several other infectious agents, and diagnosis depends upon sero­logic confirmation. Illness begins 2 to 4 weeks after exposure, with the incubation period possibly vary­ing with the infecting dose. There is sudden onset of fever, sweating, shivering, rigors, malaise, joint and limb pains, very severe frontal headache, retro- orbital pain and photophobia, and a mild nonproduc­tive cough. Rash is uncommon. Untreated fever usu­ally lasts 1 to 2 weeks, but pneumonia may persist for several weeks. Abnormal liver function tests are usual, but jaundice is less common. More than half of patients show “ground glass” infiltration due to pneumonitis on chest X-ray, even though severe res­piratory symptoms are unusual. Rarer complica­tions of acute Q fever are meningoencephalitis, cere­bellar ataxia, coma, myocarditis, pericarditis, infil­tration of the bone marrow by granulomas leading to bone marrow failure, orchitis, and placentitis. There may be splenomegaly and lymphocytosis.

Liver biopsy reveals small granulomas containing fibrinoid material. In animal models, coxiellae per­sist for long periods in liver, spleen, and lymph nodes and may multiply in the placenta during pregnancy and so be excreted in vast numbers during parturi­tion. There is some evidence that this may occur also in humans.

Chronic Q fever is usually considered a rare occur­rence, particularly, but not exclusively, affecting pa­tients with preexisting aortic or mitral valve mal­formations or disease and occurring from several months up to many years after acute infection.

Acute Q fever can be treated successfully with tetracyclines, chloramphenicol, and erythromycin, but chronic Q fever is difficult to treat. Prolonged administration of combinations of tetracyclines, ri­fampicin, Iincamycin, clindamycin, and cotrimox- azole have been recommended, but eventually heart valve replacement may be unavoidable. Reinfection of prosthetic valves has been described, possibly occurring as a result of persistent extracardiac infection.

History and Geography

In 1935, an outbreak of a febrile illness among meat workers in Brisbane, Australia, was investigated by E. H. Derrick, director of the Laboratory OfMicrobiol- ogy and Pathology, Queensland Health Department, at the request of the director-general of Health and Medical Services for Queensland. In 1937, Derrick published the first report of a new disease which he called Q fever “until further knowledge should allow a better name.” He went on to show that guinea pigs could be infected by inoculating blood or urine from febrile patients, and that extracts of guinea pig liver and spleen could transmit infection to other guinea pigs. In collaboration with Derrick, F. M. Burnet and Mavis Freeman (1937) searched for a virus in ex­tracts of guinea pig liver using animal inoculation experiments and identified a rickettsia-like agent that was called Rickettsia burnetii. In the United States, G. E. Davis and H. R. Cox (1938) reported the isolation of an organism from Dermacentor ander- soni ticks collected at Nine Mile Creek in Montana in 1935, which they called Rickettsia diaporica be­cause of the ability of the agent to pass through bacterial filters. It was later proved to be identical with R. burnetii. Its infectivity for humans was un­happily demonstrated by laboratory-acquired infec­tions. Cox showed that the Q fever agent differed significantly from other rickettsias, and it has been placed in a separate genus, Coxiella.

The most significant aspects of the history of Q fever, following its discovery, were the discovery of the importance of the domestic animal cycle, the recognition of the military significance of Q fever, the high risk for laboratory workers, and the develop­ment of vaccines.

In his original paper, Derrick speculated that transmission was due to blood-sucking insects and that the organism had an animal reservoir. He iden­tified the bandicoot as a reservoir of infection. How­ever, the wildlife cycle initially suspected by Derrick has not proved to be a significant source of human infection. The domestic animal cycle is far more im­portant and explained the epidemic of Q fever dur­ing and after World War II when thousands of troops in the Balkans and Italy were infected.

The epidemic in question began in southern Yugo­slavia in March 1941, and involved more than 600 troops. Further outbreaks occurred in northern Yu­goslavia, the Crimea, Greece, the Ukraine, and Cor­sica. A German physician, H. Dennig (1947), de­scribed more than 1,000 cases, half of whom had pneumonia. Risk of illness was associated with sleep­ing on hay or straw. Further outbreaks in 1942, 1943, and 1944 were noted in German troops ex­posed to sheep and goat flocks. Following the Allied invasion of Italy, Allied troops shared the Q fever epidemic. From February to March 1945, 511 cases of atypical pneumonia were admitted to two military hospitals in Naples. Other outbreaks occurred in Italy, Greece, and Corsica.

The military significance of Q fever was now clear. The attack rate among exposed troops was very high and illness was prolonged. For example, of 160 men billeting in a hay barn, 53 (or 33 percent) developed Q fever. Of 900 men on a training exercise, which included sitting in a bam loft to watch training films, 267 (or 30 percent) became ill. From December 1944 to June 1945, more than 1,700 cases were diag­nosed in Allied forces. After the war, outbreaks oc­curred from 1946 to 1956 among Greek and Swiss troops, in 1951 among U.S. servicemen in Libya, and in 1955 among French soldiers in Algeria. The fac­tors contributing to the epidemic among troops were the sudden exposure of susceptible people to infec­tion, particularly by sleeping on straw and hay (the local indigenous population was apparently unaf­fected by the epidemic), the rapid movement of in­fected flocks during wartime, and the mixing of in­fected and uninfected flocks. This was well illus­trated in 1974 and 1975 in Cyprus. An outbreak of abortion affected sheep and goats from flocks herded together to escape the Turkish invasion. In a camp near the place where the flocks grazed, 78 British soldiers developed Q fever.

The importance of domestic animals as the reser­voir for human infection was further emphasized by outbreaks in the United States, where the first natu­rally acquired U.S. cases were reported in 1941 in two patients from western Montana who had not received tick bites. In March 1946, 55 of 132 live­stock and meat handlers in a plant in Amarillo, Texas, developed Q fever, and two died. In August of the same year, a similar outbreak occurred in Chi­cago. Further studies revealed sporadic cases, par­ticularly in California. During outbreaks that oc­curred in 1947, city residents living near dairy farms were infected, probably by dust carried in the wind. In northern California, cases were associated with sheep. Coxiella was isolated from cow and sheep milk as well as from air and dust.

Laboratory-acquired infection was one of the earli­est problems caused by Q fever. During 1938-55, 22 incidents involving over 250 cases were recorded. In one military institution, 50 cases were recorded dur­ing 1950-65. In 1940 an outbreak occurred in the U.S. National Institutes of Health in which 15 work­ers were infected. In this and other laboratory- associated incidents, infections occurred not only in those directly handling the organism but also in those who worked near or walked through the Iabora- tory buildings, an observation that emphasized the importance of the respiratory route of infection. In the 1970s, attention was drawn to the serious risk of working on infected animals in research institutions, and interest in developing a safer and more effective vaccine was restimulated in the United States. In the 1980s, in both the United States and the United King­dom, large outbreaks occurred that were associated with operative procedures on pregnant sheep.

The first vaccines for Q fever were developed prin­cipally to protect laboratory workers from infection. Used in the late 1940s and early 1950s, they had the drawback of provoking severe local skin reactions. In the 1950s, M. G. P. Stoker and P. Fiset (1956) reported the unique phenomenon of phase variation of C. burnetii (see Immunology section), which has proved of great value not only in diagnosing chronic Q fever but also in developing an effective vaccine. The first killed vaccines used were predominantly phase 2 antigens. In the 1970s in Australia, an up­surge of infection in abattoir workers as a result of a high number of feral goats slaughtered for export led to renewed efforts in vaccine development. Vaccines currently undergoing trials in Australia are derived from phase 1 organisms.

Following the original work in Australia, the United States, and the Mediterranean area, many other countries quickly identified cases. The first re­corded human case of Q fever in the United Kingdom occurred in 1949, but seroprevalence surveys of do­mestic animals in 1952 showed infection to be wide­spread, with about 6 percent of dairy farms and 2 percent of cattle and sheep infected. These were sta­tistics that did not significantly differ from surveys done in the 1970s. By 1953, 23 countries were known to be infected, and further surveys sponsored by the World Health Organization revealed that by 1956,51 countries in all continents were infected. Only Scan­dinavia, Ireland, New Zealand, and the Netherlands were said to be free of infection. Since then, in Ireland there is good evidence that infection was introduced through importation of infected sheep from England in the 1960s, with the first indigenous cases being identified, retrospectively, in 1962. Today Q fever is known to be global in distribution, and it is likely that any patchiness in distribution reflects as much the differing levels of awareness of the disease as dif­ferences in disease incidence.

S. R. Palmer

Bibliography

American Journal of Hygiene. 1946. Reports. Paper on Q fever. 44: 1-182.

Baca, O. G., and D. Paretsky. 1983. Q fever and Coxiella burnetii: A model for host-parasite interactions. Mi­crobiological Reviews 47:127-49.

Bumet, F. M., and M. Freeman. 1937. Experimental stud­ies on the virus of “Q” fever. Medical Journal of Aus­tralia 2: 299-305.

Binuet, M. 1967. Derrick and the story of Q fever. Medical Journal OfAustralia 54: 1067—8.

Davis, G. E., and H. R. Cox. 1938. A filter-passing infec­tious agent isolated from ticks. 1. Isolation from Dermacentor andersoni, reactions in animals, and fil­tration experiments. Public Health Reports 53: 2259­67.

Dennig, H. 1947. Q fever (Balkan influenza). Deutsche medizinische Wochenschnft 72: 369—71.

Derrick, E. H. 1937. Q fever, a new fever entity: Clinical features, diagnosis and laboratory investigation. Med­ical Journal of Australia 2: 281—99.

Kaplan, M. M., and P. Bertagna. 1955. The geographical distribution of Q fever. Bulletin of the World Health Organization 13: 829—60.

Leedom₁ J. M. 1980. Q fever: An update. In Current clini­cal topics in infectious disease. Vol. I, ed. J. S. Reming­ton and M. N. Swartz, 304—13. New York.

Marmion, B. P. 1967. Development of Q fever vaccines 1937 to 1967. Medical Journal OfAustralia 54: 1074­8.

Palmer, S. R., and Susan Young. 1982. Q fever endo­carditis in England and Wales, 1975—81. Lancet 2: 1448-9.

Saah, A. J., and R. B. Hornick. 1985. Coxiella burnetii (Q fever). In Principles and practice of infectious diseases, 2d edition, ed. G. L. Mandell et al., 1088—90. New York.

Spicer, A. J. 1978. Military significance of Q fever: A review. Journal of the Royal Society of Medicine 71: 762-7.

Stoker, M. G. P., and P. Fiset. 1956. Phase variation of the Nine Mile and other strains of Rickettsia burnetii. Canadian Journal of Microbiology 2: 310—21.

Turck, W. P. G., et al. 1976. Chronic Q fever. Quarterly Journal of Medicine 45:193—217.

Worswick, D., and B. P. Marmion. 1985. Antibody re­sponses in acute and chronic Q fever and in subjects vaccinated against Q fever. Journal of Medical Micro­biology 19: 281—96.