Rocky Mountain spotted fever (RMSF) has obtained new status in the past two decades as the media attention to this disease has grown disproportionately to its true incidence. The problem remains, however, that even in endemic areas, the nonspecific nature of the presentation may not cause the physician to consider RMSF as a diagnostic possibility early in the course, when the critical period for intervention occurs. Although the disease historically has shown a higher incidence among men and children, women often live and work in tick-infested areas as well. The pregnant patient presents a special problem in that the illness may be easily mistaken for more common pregnancy-related diseases. This article discusses the microbiology, diagnosis, and treatment of RMSF, especially when the patient is pregnant. Pregnancy also presents special problems concerning the pharmacotherapy of the disease. A recent case of RMSF, occurring coincidentally with parturition and the puerperium, is described to illustrate the difficulty that sometimes is encountered in making the diagnosis and what may result from a delay in diagnosis.
A 22-year-old woman, gravida 2, para 1, was 35 weeks of gestation during the summer. Her mother found and removed a tick attached to her neck. Approximately 1 week later, the patient presented to her obstetrical provider with a complaint of worsening pain, erythema, and a raised mass at the site of the tick bite. The site was treated as a cellulitis with cephalexin. A few days later, she began complaining to family members of headache, malaise, and low-grade fever. On day 5 of the antibiotic therapy, she presented to the local community hospital with a fever of 102° F, arthralgias, malaise, and new onset of a rash. Family members described the rash in retrospect as small red blotches over her limbs and trunk. The patient was then at 36 weeks’ gestation and was admitted for observation for fever of unknown origin. The antibiotic was discontinued. She was discharged home the following day, but returned 1 day later in active labor. Colonization of the vagina by Group B Streptococcus (GBS) had been documented previously, but the rapid nature of her labor resulted in her not receiving GBS prophylaxis. Although the patient had a low-grade temperature during this hospitalization, she was discharged home on postpartum day 1.
The patient was readmitted on postpartum day 3 with recurrent fever and abdominal pain. A CT scan of her abdomen showed possible retained products of conception in her uterus, but no other abnormality. Blunt curettage was performed, but very little tissue was obtained. Laboratory studies from this readmission showed transaminases >1000 U/liter and platelets of 65,000 × 103 per mm3. Her coagulation studies were consistent with early disseminated intravascular coagulation (DIC). The patient also began to experience respiratory distress. She was then transferred to the medical intensive care unit (MICU) at the nearest tertiary care center. Just before transfer, she was given a first dose of doxycycline.
On admission to the MICU, the patient began to worsen and subsequently was intubated. She remained hypotensive with continued DIC, thrombocytopenia, and abnormal liver function tests results. Before intubation, the patient complained primarily of abdominal pain in the right upper quadrant. No uterine tenderness was found on examination. She was treated with broad-spectrum antibiotics, and serologies were sent for Rocky Mountain spotted fever, Lyme disease, Epstein-Barr virus (EBV), and cytomegalovirus (CMV). The results of these serologies were negative. On her second day in the MICU, the patient continued to worsen. She was hypotensive with worsening metabolic acidosis and required maximal doses of vasopressor support. The differential diagnosis included HELLP syndrome, acute fatty liver of pregnancy, or thrombotic thrombocytopenic purpura; sepsis secondary to tick-borne illness; postpartum toxic shock syndrome; Gram-negative sepsis secondary to uterine infection; and viral hepatitis with fulminant liver involvement.
The patient received plasmapheresis, but had little response. On MICU day 3, because of continued deterioration manifest by a serum lactate of 11 mmol/liter, an arterial pH of 6.9, and maximal ventilator support for adult respiratory distress syndrome (ARDS), the decision was made to perform an exploratory laparotomy to find a possible source of the continued acidosis. At surgery, no ischemia of bowel or other organs was noted. The decision was made to proceed with hysterectomy to remove the possible source of streptococcal toxic shock syndrome. Coincident with her surgery, the patient began improving. Within 3 days, she was weaned of all blood pressure support agents. Her metabolic acidosis rapidly improved. Broad-spectrum antibiotic coverage, including doxycycline, was continued throughout this period. On MICU day 6, her Rocky Mountain spotted fever titer was reported as positive with a latex agglutination titer of 1:256.
The final pathology from the uterus was negative for microabscesses. The patient continued to improve slowly, although a pelvic venous thrombus and pneumonia complicated her remaining MICU course. Eventually, she was extubated and left the hospital on postpartum day 24. The patient’s infant had transient hyperbilirubinemia, but was discharged in good condition when less than 1 week old.
Rocky Mountain spotted fever was first described in Idaho in 1896 (1). The role of the tick bite in transmission was suspected as early as 1904 (2). The tick is now known to be both the reservoir and the vector in nature for the pathogenic agent, Rickettsia rickettsii. In the United States, the common tick vectors are Dermacentor variabilis, the American dog and deer tick in the eastern two thirds of the U.S. and the far west; and D. andersoni, the Rocky Mountain wood tick, in the western and mountain states (1). RMSF follows the seasonal distribution of the tick with 90% of cases occurring from April to September (3). Approximately 600 to 1200 cases are reported each year in the United States with the highest incidence in the South Atlantic states, such as North Carolina and South Carolina (3). Children ages 5 to 9 years have the highest incidence of disease (3).
R. rickettsii are obligate intracellular bacteria. After attachment to the skin, the tick begins to transmit the rickettsia by way of its salivary glands after a minimum feeding time of 6 to 10 hours (1). In some cases, transmission may not occur for a longer period. Rickettsiae enter through the blood and lymphatic vessels in the skin and circulate. The organisms attach to and invade vascular endothelial and smooth muscle cells in many organs, including the skin, lungs, kidneys, liver, and central nervous system (4). The bacteria replicate by binary fission in the cytosol (4). Direct extension to contiguous endothelial cells may occur by propulsion into filopodia (4). In this way, foci of contiguous infected cells are established with little host response. The rickettsia may then alter the normal surface of the endothelial cells in a number of ways. The expression of certain membrane glycoproteins and platelet-binding substances may be altered in infected endothelial cells. Direct cytopathic effects occur secondary to the generation of oxygen-free radicals by the infected cell, rickettsial phospholipase A2, and rickettsial protease (4). Generalized vascular injury then occurs with endothelial permeability and activation of the coagulation cascade (1, 2, 4). Derangement in the coagulation and fibrinolytic pathways are indicated by evidence of activation of both the intrinsic and extrinsic coagulation pathways, as well as changes in the anticoagulant thrombomodulin by infected cells (4).
The incubation period from initial rickettsemia to clinical disease averages 7 days, with a range of 2 to 14 days (1). Fever, headache, and myalgias are common in the early stages. The fever usually will be greater than 102° F. Nausea, vomiting, and abdominal pain may also accompany the early proinflammatory phase. The rash usually appears 3 to 5 days after the onset of fever, and occurs in 84% to 91% of patients overall (1). The rash typically consists of 1- to 4-mm macules that blanch, classically appearing first on the wrists and ankles, and later on the trunk. The macules may later appear as petechiae or even coalesce into larger areas of necrosis (2). Involvement of the palms and soles is characteristic.
The major sequelae are related to the amount of endothelial damage. Complications may include pulmonary edema and ARDS. Skin necrosis with subsequent need for amputation of digits or limbs is described (1). Thrombocytopenia (<150 × 103/mm3) may occur in 32% to 52% of patients with RMSF (4). The mechanism may be either increased platelet adhesion to altered endothelial cells or disseminated intravascular coagulation, as occurs in 9% of cases (4). Hypoalbuminemia from capillary leak may contribute to severe hypotension. Acute neurological injury can manifest as hearing or motor deficits, aphasia, athetosis, and seizures (5). Although these deficits usually resolve, persistent impairment has been reported, including cerebellar, motor, speech and hearing deficits, peripheral neuropathy, and incontinence (5). The long-term dysfunction in patients who survive the acute illness results from these neurological injuries or from limb amputation (2, 5).
If treatment is not initiated, death may occur within 8 to 15 days from the onset of symptoms. Fulminant RMSF, seen more often in African-American men with glucose-6-phosphate dehydrogenase deficiency, may cause death as quickly as in 5 days (1). The mortality for untreated RMSF is approximately 25%(2). Mortality with the use of antibiotic therapy is approximately 5%(3, 6). A three- to four-fold higher case-fatality ratio is seen if treatment is delayed until 5 or more days from the onset of symptoms (3, 6). One report showed that only 49% of patients received the proper antibiotic therapy by day 5, although 90% of patients had seen a physician within those first 5 days (6). Such data underline the difficulty encountered in making the diagnosis in some cases.
With the nonspecific, early symptoms described by many patients, the diagnosis of RMSF may be difficult. The rash is present in only 49% of patients within the first 3 days (1). Although a tick-bite history may not be present in 15% of cases, the relation of such an event should prompt the physician to look for RMSF and other tick-borne illnesses, such as ehrlichiosis, babesiosis, or Lyme disease (7). Of these, ehrlichiosis most closely resembles RMSF. Ehrlichia are members of the Rickettsiaceae family and are similar intracellular, Gram-negative bacteria. Patients with ehrlichiosis may present 7 to 14 days after a tick bite with fever, headache, and myalgias. The rash of ehrlichiosis is distinct from that of RMSF by appearing as maculopapular and rarely affecting the palms or soles (7). Children and adults over 60 years of age are affected more frequently. Mortality from ehrlichiosis ranges from 2% to 10%. Treatment is similar to RMSF with doxycycline being the drug of choice (7).
Aside from other tick-borne illnesses, several other conditions must be differentiated from RMSF infection. The task may be complicated because RMSF is relatively uncommon in pregnant women compared with several other conditions that are often associated with pregnancy. In a review by Herbert et al. (8), a number of infectious diseases lead the list of differential diagnoses, many of which have special implications for vertical transmission or excess morbidity in pregnancy. The list includes rubella, rubeola, toxoplasmosis, leptospirosis, enteroviruses, secondary syphilis, and Epstein-Barr virus (8). Immunologic thrombocytopenic purpura also may be considered based on the low platelet counts and petechial rash. In the third trimester patient—gravida or postpartum—as in this case, careful consideration must also be given to severe preeclampsia with liver involvement and thrombocytopenia (HELLP), thrombotic thrombocytopenic purpura, and Gram-negative sepsis (Table 1). Streptococcal toxic shock syndrome (TSS) may occur in the postpartum setting and must be differentiated from RMSF (9). Hypotension, renal impairment, liver enzyme changes, ARDS, coagulopathy, thrombocytopenia, and rash are all included in the diagnostic criteria for streptococcal TSS. The rash of TSS may differ from that of RMSF by way of the characteristic desquamation without necrosis seen in TSS (9).
Serologic confirmation of the diagnosis can be performed, but may not be available until well into the convalescent period (1). Indirect fluorescent antibody (IFA) tests are widely available and are highly sensitive (94%) and specific (10). A four-fold rise of titers in paired samples using any method, or a one-time convalescent titer of ≥1:64 with either the IFA or latex agglutination test, is necessary to make the diagnosis (10). Direct immunofluorescence methods or DNA polymerase chain reaction tests can be applied to skin biopsy specimens from patients with the rash of RMSF (1). This may provide a more timely diagnosis during the rash phase. PCR techniques to detect rickettsia in the blood have not proven to be sensitive until late in the course of fatal cases (2).
There is little information on the risk of human vertical transmission of RMSF. At least one group has attempted to examine the placenta of a mother who underwent successful treatment of RMSF with chloramphenicol at 28 weeks of gestation (11). Immunoperoxidase staining and polymerase chain reaction techniques for Rickettsia species were performed on the placenta after term delivery and were negative in that case (11)). In the case described above, the diagnosis was not made until postpartum, and the placenta was discarded at the outside hospital. In murine models aimed at elucidating the natural enzootic cycle for rickettsia, transplacental passage has not been demonstrated for two typhus-group rickettsia, Orientia tsutsugamushi and R. typhi(12, 13). RMSF rarely occurs among infants, which is another indication that vertical transmission may not occur in humans (14, 15). The youngest serologically confirmed case occurred in a 6-month-old child whose older sibling also had acquired the disease in the same period (15).
Successful treatment of RMSF was first achieved with chloramphenicol and tetracycline in the late 1940s (2). Tetracycline has been associated with a lower mortality compared with chloramphenicol therapy and has become the preferred antibiotic in nonpregnant patients (2, 3). A 1995 retrospective report demonstrated a lower mortality in patients treated with tetracycline, even when controlling for hospitalization, age, and delay in initiating therapy (3). Also chloramphenicol is not available as an oral medication (14). Among the tetracyclines, doxycycline is the preferred agent because of its lower MIC (3). The recommended therapy is 100 mg twice daily for a minimum of 5 to 7 days, or until the patient is afebrile for at least 2 days (2). Empiric therapy is warranted in suspected cases of tick-borne illness because early treatment will improve the outcome (6). An added advantage of doxycycline is its efficacy against other tick-borne diseases, such as ehrlichiosis and Lyme disease (7). Fluoroquinolones may become another treatment option in the future. Ciprofloxacin has proven to be effective against R. conorii, and is effective against RMSF in dogs (2).
Previously, chloramphenicol was the recommended treatment for RMSF in patients younger than 9 years of age (14). This recommendation was based on the potential for permanent staining with incorporation of tetracycline in teeth. The risk of teeth staining is related, however, to the dose and duration of exposure and is unlikely to occur with short courses of therapy (14, 16). Chloramphenicol has an added disadvantage of causing irreversible aplastic anemia and may not be effective against ehrlichiosis. The American Academy of Pediatrics (AAP) amended its recommendations in 1991, and doxycycline is currently the preferred antibiotic in children (14).
Unfortunately, the same safety with short courses of doxycycline cannot be assumed in pregnant women. Tetracycline use during pregnancy is associated with fluorescent yellow discoloration of the teeth and with maternal hepatotoxicity (17). Doxycycline is suspected of having side effects similar to those of other tetracycline derivatives, although it has not been directly linked to a specific fetal malformation (17). However, because calcification of the permanent teeth does not begin until after birth, a short course of doxycycline during pregnancy is unlikely to result in discoloration of the adult teeth (8). Despite the lack of proven permanent harm with doxycycline use, chloramphenicol remains the recommended therapy for RMSF acquired during pregnancy (1, 8, 11). The recommended dose is 50 to 75 mg/kg per day in four divided doses (1). Teratogenicity has not been proven with chloramphenicol, but maternal side effects such as aplastic anemia or reversible bone marrow suppression can occur (8). Chloramphenicol is also associated with “gray baby” syndrome. This has been reported when the drug is administered to neonates, but theoretically could occur in the fetus or newborn whose mother has been treated. Fetal levels of the active drug may reach 30% to 80% of maternal levels (8). Based on such information regarding chloramphenicol, and the newer AAP guidelines for doxycycline use in pediatric cases, one could argue for the use of doxycycline therapy in near-term gravidas with a suspicious clinical picture for RMSF.
The best strategy is for primary prevention of RMSF. One option is to avoid tick-infested areas, such as dense woods or thick scrub areas. If exposure to such areas is unavoidable, then frequent tick checks will help identify tick attachment before the transmission of Rickettsia. As stated before, transmission usually begins after a minimum 6 hours of attachment (2). Ticks should be removed carefully with tweezers or covered fingers. Human infection can occur from exposure to tick hemolymph when removing the organism from persons or animals (1). Crushing the tick between fingers is also not advised.
The use of long clothing and hats are recommended in tick areas. Clothing may be soaked or covered with permethrins, and exposed skin may be covered with DEET-containing repellents. There are no specific warnings against the use of DEET in pregnancy. Teratogenic effects were not demonstrated in fetal rats and rabbits exposed to high doses of DEET in utero(18, 19). It is thought that only about 10% of the chemical is absorbed in humans after topical application (20). It is recommended that pregnant women limit their exposure to DEET by minimizing exposed skin with clothing coverage and that they apply repellents with no higher concentration than 35% DEET (20). Furthermore, limitation to tick exposure should occur through treatment of pets with permethrins, which are considered class B agents, or newer antiparasitic agents, such as fipronil. Prophylactic antibiotic administration after tick exposure is not recommended (2, 14). Such therapy of asymptomatic individuals may result in prolongation of the incubation period and has not been proven clinically efficacious or cost effective (2).
The classic triad of fever, headache, and characteristic rash occurring 1 to 2 weeks after a tick bite in an endemic area should raise suspicions for RMSF. All physicians with responsibility for the primary care of women should be familiar with the diagnosis and treatment of this illness. Early treatment with doxycycline or chloramphenicol is important in limiting the morbidity and mortality of RMSF. Pregnant patients pose special problems in both diagnosis and therapy. The time of year, local prevalence, and patient’s history of exposure may be taken into account when deciding to treat during pregnancy. Avoidance of tick-infested areas or the proper use of clothing and repellents, combined with frequent tick checks, will reduce the risk of infection.
1. Walker DH, Raoult D. Rickettsia rickettsii
and other spotted fever group Rickettsiae. In: Mandell GL, Bennett JE, Dolin R, eds. --j Principles and Practice of Infectious Diseases, 5th Ed. Philadelphia: Churchill Livingstone, 2000, vol --a 2, pp 2035–2042.
2. Thorner AR, Walker DH, Petri WA Jr. Rocky Mountain spotted fever. Clin Infect Dis 1998; 27: 1353–1359.
3. Dalton MJ, Clarke MJ, Holman RC et al. National surveillance for rocky mountain spotted fever, 1981–1992: Epidemiologic summary and evaluation of risk factors for fatal outcome. Am J Trop Med Hyg 1995; 52: 405–413.
4. Elghetany MT, Walker DH. Hemostatic changes in Rocky Mountain spotted fever and Mediterranean spotted fever. Am J Clin Pathol 1999; 112: 159–168.
5. Kirkland KB, Wilkinson WE, Sexton DJ. Therapeutic delay and mortality in cases of Rocky Mountain spotted fever. Clin Infect Dis 1995; 20: 1118–1121.
6. Archibald LK, Sexton DJ. Long-term sequelae of rocky mountain spotted fever. Clin Infect Dis 1995; 20: 1122–1125.
7. Taege AJ. Tick trouble: Overview of tick-borne diseases. Cleve Clin J Med 2000; 67: 241–248.
8. Herbert WNP, Seeds JW, Koontz WL et al. Rocky mountain spotted fever in pregnancy: Differential diagnosis and treatment. South Med J 1982; 75: 1063–1066.
9. Bisner AL, Stevens DL. Streptococcus pyogenes
(including streptococcal toxic shock syndrome and necrotizing fasciitis. In: Mandell GL, Bennett JE, Dolin R, eds. --j Principles and Practice of Infectious Diseases, 5th Ed. Philadelphia: Churchill Livingstone, 2000, vol --a 2, 2101–2117.
10. Kaplan JE, Schonberger LB. The sensitivity of various serologic tests in the diagnosis of Rocky Mountain spotted fever. Am J Trop Med Hyg 1986; 35: 840–844.
11. Markley KC, Levine AB, Chan Y. Rocky Mountain spotted fever in pregnancy. Obstet Gynecol 1998; 91: 860.
12. Shirai A, Chan TC, Gan E et al. Lack of transplacental infection with scrub typhus organisms in laboratory mice. Am J Trop Med Hyg 1984; 33: 285–287.
13. Arango-Jaramillo S, Wisseman CL Jr, Azad AF. Newborn rats in the murine typhus enzootic infection cycle: studies on transplacental infection and passively acquired maternal antirickettsial antibodies. Am J Trop Med Hyg 1988; 39: 391–397.
14. Abramson JS, Givner LB. Rocky Mountain spotted fever. Pediatr Infect Dis J 1999; 18: 539–540.
15. Jacobs WM. Rocky Mountain spotted fever in an infant: Diagnosis in siblings. Am J Dis Child 1978; 132: 928–929.
16. Lochary ME, Lockhart PB, Williams WT Jr. Doxycycline and staining of permanent teeth. Pediatr Infect Dis J 1998; 17: 429–431.
17. Briggs GG, Freeman RK, Yaffe SJ. Drugs in Pregnancy and Lactation, 5th Ed. Baltimore: Williams & Wilkins, 1998,
18. Wright DM, Hardin BD, Good PW et al. Reproductive and developmental toxicity of N,N-diethyl-m-toluamide in rats. Fundam Appl Toxicol 1992; 19: 33–42.
19. Schoenig GP, Neeper-Bradley TL, Fisher LC et al. Teratologic evaluations of N,N-diethyl-m-toluamide (DEET) in rats and rabbits. Fundam Appl Toxicol 1994; 23: 63–69.
20. Rose SR. Pregnancy and travel. Emerg Med Clin North Am 1997; 15: 93–111.