CHAPTER XVI
Q Fever
John H. Dingle, M.D., Sc. D.
HISTORICAL NOTE
In 1935 in Australia, Q fever was first recognized in man, and the causative agent was determined to be a rickettsia. In the same year, a rickettsia, subsequently shown to be of the same species as the Australian strains, was isolated by Davis and Cox (p. 402) from ticks of the species Dermacentor andersoni, collected along Nine Mile Creek in the State of Montana. For convenience, the history of the work in these two parts of the world will be summarized separately.
The disease in Australia.-Cases of an unusual type of febrile disease were first noted in Queensland, Australia, in 1933. They occurred with increasing frequency during subsequent years, and in 1937 Derrick1 described the illness in man and reported the transmission of a febrile infection to guinea pigs using as inocula both blood and urine from patients.
The clinical picture in man was characterized by an incubation period of 15 days or less, an acute onset, and symptoms of fever, headache, pains in the back and limbs, malaise, and anorexia. Chilly sensations, occasionally frank chills, jaundice, constipation, vomiting, and abdominal distention were also observed in some patients. Symptoms referable to the respiratory tract were seldom noted. Physical examination was ordinarily not remarkable. The patients appeared acutely, but not seriously, ill. The pulse was relatively slow in relation to the temperature. There was no rash. Only 4 of the 9 individuals with the original cases had cough or rales in the chest. The spleen, lymph nodes, and liver were not enlarged. Total and differential leukocyte counts were within normal limits; urinalysis revealed only slight albuminuria. The course was typified by fever, either sustained at a fixed point or swinging between 102? and 104? F. and falling to normal by lysis. Duration of illness varied between 7 and 24 days, and the patients improved gradually. Relapse was noted in a few instances. By 1940, 3 deaths had been reported among 145 cases.2
1Derrick, E. H.: "Q" Fever, a New Fever Entity: Clinical Features. Diagnosis, and Laboratory Investigations. M.J. Australia 2 : 281-299, 21 Aug. 1937.
2Cilento, R. W.: Annual Report on the Health and Medical Services of the State of Queensland for the Year 1939-40. Brisbane: Government Printer.
402
Clinical laboratory studies of patients' serums failed to give any evidence that the infection was due to a known bacterium, rickettsia, or virus. However, after Derrick transmitted the infection to guinea pigs, Burnet and Freeman3 determined that the agent was a rickettsia which Derrick subsequently termedRickettsia burneti. It was pathogenic for guinea pigs, mice, and monkeys and differed immunologically from other known rickettsiae. Convalescent serums from patients specifically agglutinated the organism but did not show agglutinins for Proteus OX-19 or OX-K (Weil-Felix reaction).
Extensive epidemiological studies were carried out by the Australian workers in an attempt to elucidate the mode of transmission of the disease. The great majority of the human cases were in abattoir employees, and the remainder were principally dairy and forest workers. Investigation of arthropod vectors showed that certain species of ticks harbored the rickettsia and that others could be infected experimentally. Similarly, bush animals, such as the bandicoot, cows, and dogs, were found either to be susceptible to infection or to be naturally infected. Derrick postulated a basic cycle of infection involving the bandicoot and other bush animals with the ticks Haemaphysalis humerosa and probably Ixodes holocyclus as vectors. Cattle might interrupt this cycle from the bite of I. holocyclus, and a secondary cycle in cattle might be established with Haemaphysalis bispinosa as the vector. Human infection might occur from either cycle but was considered more likely from the secondary cattle cycle.
This hypothesis had several gaps, however, and was presented with qualifications. With respect to human infection in particular, there was little or no indication that tick bites were responsible. Since tick feces were known to contain large numbers of rickettsiae, it was suggested that inhalation of dried tick feces from the hides of cattle might be the mode of infection of abattoir workers. Moreover, tick transmission did not explain the infection of laboratory workers that occurred in the course of this work. Both direct contact with infected tissue, and transmission by a mite were suspected but not proved.
The disease in America.-In contrast to developments in Australia, workon the disease in America began with the isolation of the etiological agentfrom ticks approximately 3 years before the first human case was recognized.
In 1935, at the Rocky Mountain Laboratory, U.S. Public Health Service,Hamilton, Mont.,4 Davis and Cox isolated a rickettsialike organism fromwood ticks of the species D. andersoni collected in Montana and later inWyoming. The agent infected guinea pigs, rats, and mice; was transmitted
3(1) Burnet, F. M., and Freeman, M.: Experimental Studies on the Virus of "Q" Fever. M.J. Australia 2: 299-305, 21 Aug. 1937. (2) Burnet, F. M., and Freeman, M.: The Rickettsia of "Q" Fever ; Further Experimental Studies. M.J. Australia 1 : 296-298, 12 Feb. 1938.
4(1) Davis, G. E., and Cox, H. R.: A Filter-Passing Infectious Agent Isolated From Ticks. I. Isolation From Dermacentor andersoni, Reactions in Animals, and Filtration Experiments. Pub. Health Rep. 53 : 2259-2267, 30 Dec. 1938. (2) Davis, G. E.: Rickettsia diaporica : Recovery of Three Strains From Dermacentor andersoni Collected inSoutheastern Wyoming : Their Identity With Montana Strain 1. Pub. Health Rep. 54: 2219-2227, 15 Dec. 1939.
403
by ticks and passed through the eggs of infected female ticks; was filterable; and grew only in tissue culture and chick embryos. The name "Rickettsia diaporica" was proposed for this organism.
During the course of this work at Hamilton, a staff member from the National Institute of Health, U.S. Public Health Service, Bethesda, Md., visited the laboratory and worked there for a few days. Subsequently, he developed an illness very similar to Australian Q fever.5 From his blood was isolated a rickettsia similar to that being studied in the Hamilton laboratory. The manner of infection could not be determined. This was the first recognized case of Q fever in the United States.
Subsequently, evidence was obtained of natural infection of man in the northwestern part of the United States by isolation of the agent from the blood or by serologic means. Although the rickettsia had been isolated from several species of ticks collected in various parts of the United States, little was known about animal hosts or the manner of spread of the infection in nature. Experimental transmission in the laboratory was possible, however, with additional species of ticks.
In the spring of 1940, the first outbreak of Q fever in the United States occurred among the 153 employees occupying one building at the National Institute of Health.6 At least 15 cases occurred with one death due to the disease. The diagnosis was confirmed in most of the cases either by isolation of the rickettsia or by serologic methods. Clinically, the study of these cases differed from that of Australian Q fever cases in one important respect; namely, that roentgenographic examinations of the chest were made. Despite the fact that symptoms and signs referable to the lungs were minor or lacking in most of the patients, the X-ray films of 15 of them showed either single or multiple, soft, infiltrating pulmonary lesions. The histopathological picture of the lungs of the individual with the fatal case was one of scattered, small, patchy peribronchial pneumonic areas with purulent exudate in the bronchi and focal necrosis of bronchial and bronchiolar walls. The similarity to primary atypical pneumonia was obvious, and it was for this reason that Q fever was subsequently considered in the differential diagnosis of the cases of atypical pneumonia reported from various parts of the United States at that time and subsequently.
The manner of spread of the infectious agent in the outbreak of the disease at the National Institute of Health was not satisfactorily determined. No cases occurred in the wing of the building where the Q fever work was
5Dyer, R. E. : A Filter-Passing Infectious Agent Isolated From Ticks. IV. Human Infection. Pub. Health Rep. 53: 2277-2282, 30 Dec. 1938.
6(1) Hornibrook, J. W., and Nelson, K. R.: An Institutional Outbreak of Pneumonitis. I. Epidemiological and Clinical Studies. Pub. Health Rep. 55: 1936-1944, 25 Oct. 1940. (2) Dyer, R. E., Topping, N. H., and Bengtson, I. A.: An Institutional Outbreak of Pneumonitis. II. Isolation and Identification of Causative Agent. Pub. Health Rep. 55: 1945-1954, 25 Oct. 1940. (3) Lillie, R. D., Perrin, T. L., and Armstrong, C.: An Institutional Outbreak of Pneumonitis. III. Histopathology in Man and Rhesus Monkeys in the Pneumonitis Due to the Virus of "Q" Fever. Pub. Health Rep. 55 : 149-155, 24 Jan. 1941.?
404
being carried on. The sex ratio of the individuals with cases was about the same as that of the employees. There was no evidence of arthropod transmission, and no secondary cases occurred in families. In retrospect, it seems probable that airborne transmission was responsible and that those persons who had been working with Q fever had already acquired inapparent infections and were immune.
Comparative studies of the Australian and American strains.-Comparisons of the Australian and American strains of Q fever Rickettsiae were carried out in both countries in considerable detail and with similar results. All of the strains were similar and were distinct from other known Rickettsiae. The Australian and American strains were immunologically identical. The animal virulence of the American strains appeared to be slightly greater but showed no qualitative differences. It was concluded that both strains belonged to the same species of Rickettsia for which Philip7 in 1943 proposed the name "Coxiella burnetii."
EXPERIENCE IN WORLD WAR II
At the time of the entry of the United States into World War II in December 1941, Q fever was known to relatively few American physicians and was considered by them to be a rare, exotic disease of little or no military importance. Human infection was thought to occur chiefly in Australia where animals also were infected. Although sporadic human cases were known to occur in the Northwestern United States, these cases were considered to be incidental in the tick cycle of transmission of the infection. It was known that the disease in man, acquired in laboratory outbreaks and, perhaps, naturally, might resemble primary atypical pneumonia and that Q fever should therefore be considered in the differential diagnosis of the atypical pneumonias. Knowledge of the epidemiology was uncertain, but tick transmission was considered to be the most likely mode of transmission. No specific treatment was known. Preventive or control measures were likewise unknown, although Cox8 had proposed a vaccine that was effective in guinea pigs. Laboratory tests for confirmation of the diagnosis, using the specific rickettsial antigen, had been described.
Q fever was not believed to constitute an important problem during the early phases of the war, and consideration of this disease by investigators during this period was primarily on the basis of excluding it as a possible cause of outbreaks of primary atypical pneumonia. Although rickettsial infection was suspected of having been responsible for certain sporadic cases of pneumonia, Q fever was not associated with large outbreaks of primary atypical pneumonia such as those which occurred at Camp Claiborne, La., in the winter of 1941-42.
7Philip, C. B. : Nomenclature of the Pathogenic Rickettsiae. Am. J. Hyg. 37: 301-309, May 1943.
8Cox, H. R.: Rickettsia diaporica and American Q Fever. Am. J. Trop. Med. 20 : 463-469, July 1940.
405
Recognition of the Disease
Q fever was first encountered, and recognized as such, during the latter part of World War II, when the disease occurred endemically and epidemically in the Mediterranean area and sporadically in Panama. Investigative work dealing with the disease was carried out by members of various U.S. Army medical units (notably, the 15th Medical General Laboratory at Naples, Italy, and the Office of the Chief of Preventive Medicine of the Mediterranean theater) and later by members of the staffs of the Medical Department Professional Service Schools, Army Medical Center, Washington, D.C., and of the National Institute of Health. The Commission on Acute Respiratory Diseases, Army Epidemiological Board, Preventive Medicine Service, Office of the Surgeon General, also conducted investigations with respect to this disease.
In December 1944, Maj. (later Lt. Col.) John H. Dingle, MC, while on a mission overseas in connection with activities of the Preventive Medicine Service, was informed by Maj. Gen. A. G. Biggam, RAMC, and members of his staff in the War Office in London, that outbreaks of atypical pneumonia had occurred in various units of British troops stationed in the Mediterranean theater. As described, these outbreaks had certain features which appeared to differ from outbreaks of primary atypical pneumonia in the United States in that they were localized to particular military units, were characterized by attack rates as high as 50 percent, and showed an absence of an associated increase in minor respiratory illness. Subsequently, it was learned from U.S. Army medical officers in the Mediterranean theater that the majority of cases of primary atypical pneumonia seen among U.S. troops in the theater differed from those in the United States in that they were milder clinically, showed less pulmonary infiltration roentgenographically, had a greater tendency to occur in local outbreaks, and failed to develop cold hemagglutinins. In addition, there was some evidence that the pathology was different. In the examination of the lungs of nine fatal cases, Lt. Col. Tracy B. Mallory, MC, of the 15th Medical General Laboratory, noticed the absence of the acute necrotizing bronchiolitis and bronchitis which had been considered to be characteristic of the pathology of primary atypical pneumonia.9
In February 1945, an opportunity arose to investigate one of these outbreaks in British troops in the Mediterranean theater. At the request of Brig. E. R. Boland, RAMC, Consulting Physician, C.M.F., and with the approval and cooperation of the Office of the Surgeon, MTOUSA (Mediterranean Theater of Operations, U.S. Army), an outbreak in the British 6th Parachute Battalion was studied by Lt. Col. Ross L. Gauld, MC, Major Dingle, and Capt. (later Maj.) Frederick C. Robbins, MC. The infection in this
9Letter, Maj. J. H. Dingle, MC, Respiratory Diseases Commission Laboratory, to The Surgeon General, U.S. Army, War Department, Washington, D. C., 1 June 1945, subject : Report of an Investigation of Influenza and Other Respiratory Diseases in the European and Mediterranean Theaters of Operation.
406
organization apparently had been acquired in Athens, Greece. Although the outbreak was subsiding and many of the patients were convalescing at the time of the investigation, the data obtained suggested that the disease differed from primary atypical pneumonia. Subsequently, the disease was identified as Q fever by the laboratory examination of serums from patients and other men in the organization.
During the latter part of February, March, and April 1945, Captain Robbins, and Capt. (later Maj.) F. B. Warner, MC, of the 15th Medical General Laboratory, and Maj. (later Lt. Col.) C. A. Ragan, MC, of the 15th Field Hospital, investigated four similar outbreaks in U.S. troops stationed in northern Italy near Florence and Bologna. Another outbreak, not studied in detail, occurred in troops stationed near Lake Garda. Captain Robbins and his associates at the 15th Medical General Laboratory isolated a rickettsia, subsequently identified as C. burnetii, from the blood of some of these patients and demonstrated further that this agent was the cause of the outbreak. In addition, 29 of 49 sporadic cases of an illness diagnosed as atypical pneumonia in Italy were shown actually to be Q fever.10
At the time the above studies were begun, in February 1945, Maj. (later Lt. Col.) C. J. D. Zarafonetis, MC, of the United States of America Typhus Commission, learned from Dr. J. Caminopetros of the Pasteur Institute of Greece that an outbreak resembling, but not identical with, influenza had occurred in Athens and its suburbs during the preceding winter. Using the blood of one of these patients, Dr. Caminopetros had established a febrile disease in guinea pigs which was then maintained by guinea-pig passage for the subsequent 13 months. With the approval of Dr. Caminopetros, Major Zarafonetis reported this work and forwarded specimens of infected guinea-pig blood to Brig. Gen. Stanhope Bayne-Jones of the Office of the Surgeon General for study in the United States. The specimens were sent to the Respiratory Diseases Commission Laboratory11 at Fort Bragg, N.C., where the agent was identified as C. burnetii and termed the Balkan grippe strain.
In August 1945, Col. Garnett Cheney, MC, and Maj. (later Lt. Col.) W. A. Geib, MC, isolated in guinea pigs a rickettsia from the blood of a patient with an illness diagnosed as atypical bronchopneumonia in Panama. This agent was identified at the National Institute of Health as a strain of C. burnetii.
In May and June 1945, an outbreak of respiratory disease occurred among troops in transit from Grottaglie in southern Italy to Camp Patrick Henry, Va., and other ports of debarkation in the United States. Investigation of
10(1) Robbins, F. C., and Ragan, C. A.: Q Fever in the Mediterranean Area : Report of Its Occurrence in Allied Troops. I. Clinical Features of the Disease. Am. J. Hyg. 44: 6-22, July 1946. (2) Robbins, F. C., Gauld, R. L., and Warner, F. B.: Q Fever in the Mediterranean Area : Report of Its Occurrence in Allied Troops. II. Epidemiology. Am. J. Hyg. 44 : 23-50, July 1946.
11The professional staff of the Respiratory Diseases Commission Laboratory was as follows : Maj. John H. Dingle, MC, Director ; Maj. Theodore J. Abernethy, MC ; Maj. George F. Badger, MC; Maj. Norman L. Cressy, MC ; A. E. Feller, M.D. ; Irving Gordon, M.D. ; Maj. Alexander D. Langmuir, MC ; Charles H. Rammelkamp, Jr., M.D. ; and Capt. Elias Strauss, MC.
407
the cases at Camp Patrick Henry and of other aspects of the outbreak was carried out by Maj. Marcus A. Feinstein, MC, Capt. Raymond Yesner, MC, and Maj. Jerome L. Marks, MC, of the station hospital and by members of the staff of the Respiratory Diseases Commission Laboratory. Roentgenograms of the chest were used in the study of this outbreak as a method of case findings. Although rickettsiae could not be isolated from these patients, the disease was identified serologically as Q fever.
Laboratory outbreaks of Q fever occurred among the personnel of the 15th Medical General Laboratory and of the Respiratory Diseases Commission Laboratory during the course of these investigations. The agents responsible were shown to be the so-called Italian and the Balkan grippe strains of C. burnetii, respectively.
Thus was Q fever recognized and identified as the cause of sporadic and epidemic cases of atypical pneumonia in the Mediterranean area and in Panama. In the process, new strains of C. burnetii were obtained and subsequently characterized. In all probability, Q fever occurred in U.S. troops in other parts of the world during World War II but was not definitively recognized.
Occurrence in Military Forces
Statistical data regarding the occurrence of Q fever in U.S. military forces during the war are nonexistent, probably for the following reasons: Until the latter part of the war, the disease was not recognized, if indeed it occurred, and therefore was not suspected; differentiation of this disease from primary atypical pneumonia clinically was difficult, if not impossible, and laboratory confirmation was required; and it occurred sporadically or in focal outbreaks in limited geographic areas. The most that can be stated in retrospect is that Q fever in northern Italy may have accounted for as many as 75 percent of the cases diagnosed as atypical pneumonia in certain Army hospitals and that the total attack rates in military units experiencing focal outbreaks reached levels as high as 20 or 30 percent of their strength.
Data regarding the true occurrence of Q fever are too inadequate to permit a determination of the extent to which this disease constituted a military problem during the war. It is reasonable to believe that this infection was the cause of considerable illness and loss of time among the troops in the Mediterranean area. Only one unit in Italy, the 3d Battalion, 362d Infantry Regiment, is known to have suffered an outbreak while in combat.12 The epidemic began 4 days after the unit of approximately 900 men had moved back into the line from the rest area. In the next 3 weeks, 269 men, or almost 30 percent of the unit, were hospitalized. The potential military importance of this disease in its epidemic form is thus apparent.
12See footnote 10 (2), p. 406.
408
Knowledge Acquired Through the War
The several studies of Q fever during the war so extended knowledge of the behavior and extent of this disease that it could no longer be considered as a medical curiosity. Careful analyses of the clinical aspects of the naturally occurring cases, as well as of laboratory-acquired infections, confirmed the prewar descriptions and the extremely low mortality.13 In addition, these studies emphasized both the variation that may occur in clinical severity and the high frequency of pulmonary infiltration.
With respect to the epidemiology of Q fever, the studies and observations made during World War II added several geographic areas to the known distribution of the disease. These were principally the countries of Italy and Greece and the island of Corsica in the northern Mediterranean region and Panama in Central America. Supportive evidence was obtained that the disease had occurred endemically and epidemically in various parts of Bulgaria and Greece since 1941. The endemic and epidemic occurrence of Q fever in Italy and its sporadic occurrence in Panama were clearly established.
The source of infection and mode of transmission of the naturally occurring disease were not determined. Neither animal reservoirs nor arthropod vectors were clearly associated with the epidemics. In many instances, the outbreaks were sufficiently explosive to suggest a single exposure and a common source such as contaminated food or water. No evidence was obtained to support such an explanation. All of the epidemics were focal in character, and some were associated with particular places, such as barns or houses, or with straw or hay, yet the manner of transmission could not be found. There was no evidence pointing to spread of the disease by person-to-person contact. The inhalation of infected dust or other particles could not be excluded and remained as an attractive possibility, particularly by analogy to outbreaks in the laboratory, where transmission was presumably by air.
Measures for prevention and control of Q fever were not applied and, indeed, were not available because of lack of knowledge of the natural history of the disease. Isolation of cases and quarantine of contacts were not indicated because the disease seldom spread from person to person. Since the mode of spread was unknown, attempts to interrupt or break the chain of transmission could not be made. Immunization of guinea pigs with a rickettsial vaccine had been accomplished, but no attempts to immunize human beings had been made. Immunization of troops against Q fever was not attempted.
Laboratory confirmation of the diagnosis of Q fever was possible by isolation of the rickettsia in animals or embryonated eggs and by serologic
13Robbins, F. C.: Q Fever, Clinical Features. In Rickettsial Diseases of Man. Washington : Am. Assoc. Advancement Sc., 1948, pp. 160-168.
409
tests. Improved techniques were developed,14 and the use of the highly reactive Italian strain (Henzerling) as an antigen for serologic tests greatly facilitated such work.15
SUMMARY AND EVALUATION OF THE FUTURE IMPORTANCEOF Q FEVER
During World War II, the status of Q fever changed from that of a little-known, possibly exotic disease to that of a disease of considerable importance in certain highly endemic areas. The investigations of Q fever carried out by members of various U.S. Army medical units and by others extended considerably the knowledge of the clinical behavior of the disease, particularly with respect to the high frequency of pulmonary involvement. However, no effective method of therapy was found. Two widely separated geographic areas were added to the known distribution of Q fever; namely, the northern Mediterranean region and Panama. Laboratory methods for confirming the clinical diagnosis were improved, particularly by the discovery of strains of C. burnetii which were more highly reactive antigenically than those previously known. The reservoir of infection and the mode of transmission were not defined, nor were measures of prevention and control developed.
The years immediately following World War II witnessed a great extension of knowledge regarding Q fever from which it is apparent that this disease is one of considerable potential military importance. Definitive or presumptive evidence has been presented of the existence of the disease in Australia, the United States, Panama, Mexico, England, Wales, Spain, Portugal, Switzerland, France, Germany, Rumania, Yugoslavia, Greece, Turkey, Iraq, Israel, Morocco, Algeria, Libya, the Belgian Congo, French Equatorial Africa, the Union of South Africa, India, and China. Possible reservoirs and hosts include at least 17 species of ticks, the body louse, cow, sheep, goat, bandicoot, and pigeon. The milk and the placentae of infected cows, sheep, and goats may be heavily contaminated; the urine and feces also of these animals may occasionally be heavily contaminated. Thus, the environment of infected domestic animals may become heavily contaminated, and the contamination may persist for weeks because of the extraordinary resistance of C. burnetii to drying and to heat.
14(1) Bengtson, I. A. : Complement Fixation in the Rickettsial Diseases : Technique of the Test. Pub. Health Rep. 59: 402-405, 24 Mar. 1944. (2) Gallenson, N.: Hypertonic Sodium Chloride Solution as Serum Diluent in Agglutination Tests With Rickettsia Burneti. Proc. Soc. Exper. Biol. & Med. 63: 169-171, October 1946.
15(1) Robbins, F. C., Rustigian, R., Snyder, M. J., and Smadel, J. R. : Q Fever in the Mediterranean Area : Report of Its Occurrence in Allied Troops. III. The Etiological Agent. Am. J. Hyg. 44 : 51-63, July 1946. (2) Topping, N. H., Shepard, C. C., and Huebner, R. J.: Q Fever : An Immunological Comparison of Strains. Am. J. Hyg. 44 : 173-182, July 1946.
410
Although many questions remain regarding the epidemiology of Q fever, it now appears that most human infections arise in association with animals and their environment, some arise from the drinking of raw milk, and rare cases may follow the bite of an arthropod vector or patient-to-patient transmission. The respiratory tract appears to be the most common portal of entry in man, and the inhalation of contaminated dust probably is the mode of infection.
Effective and specific therapy can probably be obtained with one of the broad-spectrum antibiotics (Aureomycin, chloramphenicol, and Terramycin), although further evaluation is desirable because of the variability of the clinical disease. Preventive and control measures need further investigation. Pasteurization of milk and milk products is obviously indicated. Vaccination appears to have been effective on a small scale but needs evaluation on a larger scale before it can be employed militarily. The possibilities of chemoprophylaxis also require investigation.
