Colorado tick fever

Camp Bullis, Texas, is an active training facility for the U.S. Army and Air Force with a storied history dating back to the late 19th century. In the early 1940s, an epidemic of a seasonal tick-borne rickettsial-like illness occurred at Camp Bullis; the last case was reported in 1947. To date, the etiology of this disease has remained elusive. In an attempt to retrospectively identify the causal pathogen, we surveyed Camp Bullis for the presumed tick vector with intent to screen molecularly for Rickettsia and Ehrlichia. However, no ticks were recovered from primary dragging attempts in the spring or from harvested deer in the fall. Moreover, pathologic and microbiologic materials obtained during the epidemic are no longer extant, making them unavailable for analysis. In this study, we review potential circumstances that impact emerging and, in this case, vanishing infections. The etiology of Bullis fever will probably remain undetermined, and this once-emerging infection may have vanished into history. However, given Camp Bullis' status as an active medical training site, awareness of and surveillance for the disease should continue.

Historical Perspective

Bullis fever was a rickettsial-like illness that occurred in World War II troops training at Camp Bullis, Texas, 18 miles northwest of San Antonio.1 This seasonal disease, first noted in the spring and summer of 1941, was diagnosed in more than 1,000 U.S. Army soldiers, causing one death.2 The disease incidence peaked in 1943 with 485 cases, dropping to 18 cases in 1945 when the number of troops training was significantly curtailed.2-6 The last recognized case was reported in 1947.7

Bullis fever was characterized by a 7- to 10-day incubation period followed by 4 to 14 days of symptoms.1,2 Symptoms developed abruptly, with an initial chill followed by fevers and postorbital or postoccipital headaches. Patients often had marked, generalized lymphadenopathy that resolved once the fever abated. Up to 10% of the cases, generally those with more severe symptoms, had a maculopapular rash predominately involving the trunk. Multiple tick bites were commonly evident.

A constant finding among patients with Bullis fever was leukopenia with associated neutropenia occurring on or about the second or third day of symptoms. The total white blood cell count frequently dropped to ~3,000/µL and occasionally to as low as ~1,750/µL. The leukopenia gradually resolved during convalescence; however, a relative lymphocytosis persisted beyond discharge. Patients were not anemic. Several patients had trace albuminuria. The clinical reports did not record values for any liver-associated enzymes (not yet developed) or platelet counts. Although defervescence was abrupt, convalescence was protracted. One death occurred and was attributed to "agranulocytic angina and sepsis."4 Treatment was mostly supportive and consisted of codeine, aspirin, ice, rest, and fluids. Thirteen of the 47 patients managed in 1944 were treated with penicillin without apparent benefit.5 One patient failed to respond to sulfa drugs and penicillin, but appeared to improve after administration of p-aminobenzoic acid, an agent then used in the treatment of rickettsial disease.7

The presence of "small intracellular (organisms) similar in morphology to Rickettsia" in patients' blood and lymph tissue, along with epidemiological evidence supporting the Lone Star tick, Amblyomma amencanum, as the vector were suggestive of tick-borne infection.2,8 However, studies performed at that time at the Brooke General Hospital and other reference laboratories were negative for Q fever, Rocky Mountain spotted fever, endemic typhus, tularemia, and rickettsialpox.2,9 Additional studies for brucellosis, typhoid fever, malaria, Epstein-Barr virus, Colorado tick fever, dengue fever, and lymphocytic choriomeningitis virus were also nondiagnostic.

Additional support for the existence of a distinct, tick-borne, rickettsial-like organism as the cause of Bullis fever was provided by extensive animal, insect, and human studies conducted in the 1940s.2,9,10 A "rickettsial-like" organism was observed in emulsified A. amencanum ticks and in guinea pigs inoculated with blood or lymph glands from patients with Bullis fever. Moreover, human volunteers developed a clinical illness consistent with Bullis fever after challenge with whole blood isolates from Bullis fever patients, emulsions from A. americanum ticks, or from human or tick strains of the presumptive Bullis fever organism propagated in chick embryos or mice. Despite these suggestive studies, however, analysis of more than 10,000 ticks and 2,500 mites from Camp Bullis revealed no contemporarily known pathogen in such frequency that Bullis fever could be ascribed to it.

The Army drastically reduced the number of troops training at Camp Bullis by 1944 because of this disease, a process that accelerated with the military demobilization after 1945.3 Consequently, the incidence of Bullis fever plummeted, with no reported case evident after 1947. Concurrently, a major effort to eliminate ticks from infected areas of Camp Bullis with dichloridphenyltrichloroethane was undertaken.3 The disease was no longer considered a risk to the welfare of troops and because the presumptively identified agent of Bullis fever could not be further characterized by the contemporary techniques, investigations ended.

Training intensified at Camp Bullis approximately 10 years later. In the interim, cases of Bullis fever were not observed in the few assigned caretaker military personnel. Tick and wildlife surveys in 1978 showed no evidence of brucellosis, tularemia, or Rocky Mountain spotted fever. More recent tick surveys performed in 1987 failed to reveal Lyme disease or Rocky Mountain spotted fever (J. Longfield, unpublished data).

Other pathogens have since been proposed as potential agents of Bullis fever, and new technologies have invited retrospective investigation. For example, polymerase chain reaction amplification with primers specific for Ehrlichia chaffeensis against DNA extracts from 106 ticks collected and preserved during the outbreak proved negative (C. Paddock and J. Goddard, unpublished data). However, these results may be ambiguous because the ticks had been preserved in ethanol for more than 50 years and there was evidence of DNA degradation.

Although Camp Bullis was used as a combat training post when it initially opened in 1890, since the 1970s it has been used primarily for training medical personnel. Camp Bullis is the premier training site for the Army Medical Department Center and School. Almost universally, every person in the Army Medical Department, and many from the Navy and Air Force medical services, undergo training at Camp Bullis. Although no cases of Bullis fever have been reported recently, an assessment was undertaken to determine what the original agent responsible for the disease was and if the possibility of transmission might persist today.

Current Investigations

To determine whether biologic samples were still available, we contacted the laboratories involved in the original investigation in the 1940s. Unfortunately, none of these institutions (University of Texas Medical Branch, Galveston, Texas; Texas Department of Health; Rocky Mountain Laboratory, U.S. Public Health Service, Hamilton, Montana; and our own institution, Brooke Army Medical Center (formerly Brooke General Hospital], Fort Sam Houston, Texas) had retained any specimens for the 50 years since they were collected. The Armed Forces Institute of Pathology (Washington, DC), which was recorded as having been sent autopsy specimens from the single soldier who died from the disease, in fact never received the specimens (G. Wagner, personal communication). We were also unsuccessful in obtaining serologie evidence of previous rickettsial or ehrlichial infection from volunteers who participated in the Bullis fever challenge trials.10

Because there were no residual pathologic or microbiologie samples from the initial outbreak, we attempted to define the etiology of Bullis fever by obtaining ticks from Camp Bullis to test them for agents that could conceivably cause this syndrome. In May 2001, we dragged for ticks and used dry ice traps in five geographically distinct areas at Camp Bullis between dawn and late morning. May was the month of the original epidemic when transmission was the highest. Not a single tick was harvested. Given that deer appeared to have been the principle host of the tick during the Bullis fever epidemic,2 we also attempted to collect ticks from hunter-harvested deer during December 2001, early in hunting season. Of 10 deer taken from multiple sites on Camp Bullis, not a single tick was found (although the season was possibly late enough that even in south Texas, few adult ticks might be expected to be found on large mammals). Ultimately, our attempts to collect ticks from varied areas on Camp Bullis using several methods over two distinct seasons were unsuccessful.

Discussion

Bullis fever was a prototypical emerging infectious disease 60 years ago. This disease was a remarkable harbinger of other rickettsial infections, such as ehrlichiosis, that would later be cited as examples of emerging infectious diseases. Bullis fever was unique among rickettsial-like diseases because of the presence of lymphadenopathy, lymphocytosis, and a restricted geographic distribution." However, Bullis fever was also unique because of its status as a once-emerging disease that has apparently vanished.

Other infectious diseases, such as malaria, dengue fever, and diphtheria, have waxed and waned tremendously in prevalence, earning the status of emerging, diminishing, or re-emerging entities. Only smallpox has truly vanished because of human intervention.12 Variations in factors such as ecology, human demography and behavior, international travel, technology and industry, microbial adaptation, and change and breakdown of public health measures all occur with enough regularity to introduce emerging infections, but they rarely cause infections to utterly vanish.13

Camp Bullis was an underoccupied military training area until a rapid influx of troops occurred at the beginning of World War II. We suspect that this radical change in demographics, with the sudden proximation of a large human population with a presumed tick vector, was responsible for the emergence of this unique disease. We also suspect that several of the aforementioned factors then culminated in the disappearance of Bullis fever. First, the unique ability of the military to entirely abandon the area at Camp Bullis for other available training locations drastically changed demographics. second, dramatic changes in vegetation and wildlife in the area resulted from a prolonged drought that occurred in the 1950s in Texas,14 along with ongoing man-made alterations of the terrain at Camp Bullis for the reintroduction of training. Third, there has been a notable decline in the presumed tick vector population at Camp Bullis over the past six decades. Our inability to isolate a single tick with multiple attempts in 2001 stands in marked contrast to, for example, the 910 adult ticks and 1,160 nymphs collected from two deer and the 294 ticks removed from one patient during the epidemic.2 We hypothesize that the presumed tick vector, A. americanum, could not survive the triple pressures of the alteration in vegetation and wildlife,15 the appearance in Texas at that time of the fire ant, a known predator of ticks,16'17 and the use of dichloridphenyltrichloroethane.3 Finally, no illness reminiscent of Bullis fever has heretofore been described and the responsible pathogen may not have existed outside of the limited geographical area of south Texas; if so, this lack of geographical flexibility may have limited the fitness of the pathogen to survive severe stresses in this region. We suspect that it was the cumulative effect of all of these factors that was responsible for the fate of Bullis fever-the apparent disappearance of the disease. Sanguinity with this result is unwarranted, however, as has been demonstrated repeatedly with the reappearances of Ebola hemorrhagic fever in central Africa from reservoirs unknown.13 The absence of Bullis fever may merely reflect our inability to detect a very low level presence of the organism, also in reservoirs unknown. We continue our educational and surveillance efforts within our system in the anticipation of a possible resurrection of the disease.

In summary, historical accounts support the existence of a distinct rickettsial-like, tick-borne infection that was recognized as Bullis fever. The causative agent was incompletely characterized. We were unable to retrospectively implicate a pathogen by modern forensic microbiologie or pathologic techniques because of the lack of stored contemporary samples as well as the disappearance of the tick vector that probably served as the reservoir. This infection was an excellent early example of an emerging infectious disease that is also unique for having apparently vanished.

References

1. Woodland JC, Richards JT: Bullis fever (lone star fever, tick fever). JAMA 1943; 122: 1156-60.

2. Anigstein L, Anigstein D: A review of the evidence in retrospect for a rickettsial etiology in bullis fever. Tex Rep Biol Med 1975; 33: 201-11.

3. Camp Bullis: Admirably Suited to All Purposes of Military Training. A history of the Leon Springs Military Reservation, 1890-1990. Fort Sam Houston Museum, Fort Sam Houston, TX.

4. Annual Report of the Brooke General Hospital, Fort Sam Houston, TX, 1943, pp 20-4.

5. Annual Report of the Brooke General Hospital, Fort Sam Houston, TX, 1944, pp 21-2.

6. Annual Report of the Brooke General Hospital, Fort Sam Houston, TX, 1945, pp 95-8,

7. Arnold WT, Van Noate HF: Bullis fever: report of a case treated with paraaminobenzoic add. Bull US Army Med Dept 1949; 9: 218-23.

8. Livesay HR, Pollard M: Laboratory report on a clinical syndrome referred to as bullis fever. Am J Trop Med 1943; 23: 475-9.

9. Livesay HR, Pollard M: Serological studies of bullis fever. Am J Trop Med 1944; 24: 281-4.

10. Pollard M, Livesay HR, Wilson DJ, Woodland JC: Experimental studies with bullis fever. Am J Trop Med Hyg 1946; 26: 175-87.

11. EngTR, McDade JE, Fishbein DB: Was bullis fever actually Ehrlichiosis? In reply. JAMA 1988; 260: 3006-7.

12. Satcher D: Emerging infections: getting ahead of the curve. Emerg Infect Dis 1995; 1: 1-6.

13. Morse SS: Factors in the emergence of infectious disease. Emerg Infect Dis 1995; 1: 7-15.

14. Bomar GW: Heat waves and drought: legacy of a Texas summer. In: Texas Weather, pp 143-56, Ed 1. Edited by Bomar GW. Austin, TX, University of Texas Press, 1983.

15. Semtner PJ, Barker RW, Hair JA: The ecology and behavior of the lone star tick (Acarina'.Ixodidae]. II. Activity and survival in different ecological habitats. J Med Entomol 1971; 8: 710-25.

16. Burns EC, Melancon DG: Effect of imported fire ant (Hymenoptera:Formicidae) invasion on lone star tick (Acarina-.lxodidae) populations. J Med Entomol 1977; 14: 247-9.

17. Drees B: Spread of the imported fire ant. Available at http://agnews.tamu.edu/ graphics/flreants/antspred.html; accessed May 22, 2003.

Guarantor; COL David P. Dooley, MC USA

Contributors: MAJ Clinton K. Murray, MC USA; COL David P. Dooley, MC USA

Department of Medicine, Infectious Disease Service, Brooke Army Medical Center, Fort Sam Houston, TX 78216.

Reprints: MAJ Clinton K, Murray, Infectious Diseases Service, Brooke Army Medical Center, MHCE-MDI, 3851 Roger Brooke Drive, Fort Sam Houston, TX 78234-2000.

This work was presented at the 52nd Annual Conference on Diseases in Nature Transmissible to Man, June 12-13, 2002, Houston, TX.

The opinions or assertions contained herein are the personal views of the authors and should not be construed as reflecting the official positions of the Department of the Army or the Department of Defense.

This manuscript was received for review in June 2003. The revised manuscript was accepted for publication in December 2003.

Copyright Association of Military Surgeons of the United States Nov 2004
Provided by ProQuest Information and Learning Company. All rights Reserved