Keywords
Rocky Mountain spotted fever - pregnancy - tick-borne - infections - doxycycline
Rocky Mountain spotted fever (RMSF), caused by Rickettsia rickettsii, is the most common tick-borne rickettsial disease in the United States.[1] Fatal outcomes have been reported in up to 20% of untreated cases and 5% of treated
cases.[2] The diagnosis of RMSF can be clinically challenging due to nonspecific and highly
variable initial symptoms as well as limited diagnostic value of serologic testing.
The classic triad of fever, headache, and rash is only observed in 3% of patients
with RMSF in the first 3 days of illness.[1]
The diagnosis of RMSF in pregnancy poses a unique challenge as a number of the clinical
manifestations, such as myalgias, fatigue, nausea, vomiting, headache, and abdominal
and back pain are common in pregnancy, which may delay further diagnosis. Severe manifestations
such as respiratory distress, acute renal failure, and neurologic changes are associated
with other pregnancy-related diseases including preeclampsia and hemolysis, elevated
liver enzymes, and low platelet count (HELLP) syndrome.[3] Furthermore, the standard of care treatment for RMSF is doxycycline and chloramphenicol,
which were formerly classified as Class D in pregnancy and are generally avoided in
pregnancy.[1]
We describe a case of an atypical presentation of RMSF in pregnancy successfully treated
with doxycycline.
Case
A 37-year-old gravida 2, para 1 previously healthy woman was referred to our hospital
during the summer of 2019 at 310/7 weeks' gestation due to severe headaches for a week and a fever of 38.9°C in her
obstetrical office. Her pregnancy was complicated by a history of migraines, though
this headache presented differently from her typical migraine. She described this
headache as a severe frontal headache with mild photophobia and phonophobia that was
not relieved with over-the-counter medication. It began during travel to Washington
state in the prior week, most of which was spent in Seattle. However, she did note
a brief walk in the woods but reported no tick bites, signs of tick bites, nor contact
with any animals other than her own dog which showed no change in behavior. She worked
as an art curator in North Carolina. Over the few days prior to admission, she developed
night sweats, chills, sweating, dyspnea, and one episode of diarrhea.
On admission, she continued to report unremitting headache. No rash, acute meningeal
signs, evidence of lymphadenopathy, nor focal neurological deficits was noted. Her
vital signs were normal, and fetal status was reassuring. However, laboratory assessment
was significant for normal white blood cell (9.2 × 109/L) with bandemia (23%) noted on manual differential, transaminitis (aspartate transaminase
104), anemia (lactate dehydrogenase 326), hemoglobin (9.7 g/dL), proteinuria (3+ on
urine analysis with P:C ratio of 1,203 mg/g), and thrombocytopenia (76 × 109/L). No prior history of thrombocytopenia or proteinuria was evident on her prenatal
records. Infectious urine analysis and respiratory panel (including influenza) were
negative. She was acutely treated for atypical HELLP syndrome, despite being normotensive,
with betamethasone, intravenous (IV) fluids, and IV magnesium, while a further infectious
workup was pending given her fever. Haptoglobin was normal, and a manual peripheral
smear showed normal platelet counts and no evidence of schistocytes, hemolysis, or
platelet clumping. Magnesium was then discontinued, with further infectious workup
pursued, including lumbar puncture and a noncontrast brain magnetic resonance imaging
(MRI). She was started on empiric IV vancomycin and meropenem for unknown infectious
etiology.
Six hours after antibiotic initiation, her vital signs worsened with continued fever,
severe tachypnea, tachycardia, and a new oxygen requirement (89% on room air to 95%
on 3-L nasal cannula). Because of this acute decompensation, she was transferred to
the intensive care unit (ICU).
Overnight, the fetal heart rate displayed repetitive decelerations to the 70 to 80
and the patient developed respiratory distress and desaturation quickly after (respiratory
rate 30, heart rate 111, SpO2 97% on 4-L nasal cannula). The decision was made to proceed with intubation and urgent
cesarean section due to maternal and fetal distress. She tolerated the surgery well
and was transferred back to the ICU. Neonatal condition was stable with a weight of
1,775 g and Apgar scores 9 and 9. The neonate was taken to the neonatal intensive
care unit (NICU) due to gestational age.
The patient was started on doxycycline immediately after vancomycin was discontinued.
RMSF became the presumed diagnosis due to the clinical criteria of headache, fever,
and malaise, laboratory findings notable for thrombocytopenia, transaminitis, acutely
worsening disease, and cerebrospinal fluid analysis consistent with aseptic meningitis.
Tissue biopsies were not taken at the time of acute infection; however, the placenta
was sent for pathologic evaluation. Her cerebrospinal fluid studies were consistent
with aseptic meningitis. She was noted to have significant improvement with doxycycline
treatment, although her postpartum course was complicated by acute kidney injury,
thrombocytopenia, ileus, and stable transaminitis. Her brain MRI was unremarkable,
and she was extubated 3 days later with a normal neurologic examination. She was discharged
home on postoperative day 6.
The acute RMSF antibody, which resulted after her decompensation and delivery, was
negative. Microabscesses seen on pathologic evaluation of the placenta were concerning
for Listeria infection. However, placental cultures for Listeria and Rickettsia were negative. Rickettsial placental immunohistochemical (IHC) staining was performed
and was also negative. The patient's clinical syndrome and improvement on doxycycline
were consistent with RMSF; her convalescent serologies were positive (> 1:256). The
infant overall did well in the NICU and showed no evidence at delivery or in the neonatal
period of RMSF or Listeria infection.
Discussion
We present an atypical case of RMSF in the third trimester in a patient who was successfully
treated with doxycycline. Literature around RMSF during pregnancy has been limited.[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10] In endemic regions, especially eastern and south-central United States, the presence
of a fever, transaminitis, thrombocytopenia, and rash in any patient with a history
of a tick bite, pregnant or not, presumes the diagnosis, and empiric therapy should
be started. However, in the setting of pregnancy, even in an endemic area, the differential
diagnosis for fever with transaminitis and thrombocytopenia includes a number of other
infectious diseases (rubella, streptococcal toxic shock syndrome, and Epstein–Barr
virus) as well as immunologic and rheumatologic causes (immunologic thrombocytopenic
purpura). In the pregnant population, severe preeclampsia and HELLP can also present
with transaminitis and thrombocytopenia.[3]
RMSF can be promptly suspected with the presentation of a petechial rash on the palms
and soles 2 to 5 days after the onset of fever.[2]
[8] However, up to 20% of cases present with an atypical or absent rash.[1]
[2] This is the first case to our knowledge to report RMSF in pregnancy without a significant
rash or tick history. It is possible that a rash may be less likely in pregnancy,
but these data are limited. Given that fatal outcomes have been reported in up to
20% of untreated cases and 5% of treated cases, absence of a rash but appropriate
clinical picture should not exclude RMSF.[2]
The pathophysiology during pregnancy and the potential for vertical transmission are
still unknown. In a series of four cases in Mexico, RMSF infection in three patients
infected during the first trimester led to abortion, likely related to hypotension
and placental ischemia. The fourth patient was diagnosed during the third trimester
with no sequelae to either mother or fetus.[6] In one pregnancy affected by Brazilian spotted fever, autopsy results confirmed
positive IHC staining for rickettsial antigen in maternal tissue but not observed
in fetal tissues.[8]
The microabscesses in the placenta in this case posed an interesting conundrum. Though
all cultures were negative, this finding at least indicates placental inflammation.
In prior literature, pathologic evaluation of the placenta has not noted evidence
of RMSF in the placenta.[3]
[5] At this time, there is not enough evidence to suggest RMSF crosses the placenta,
and there are no reported cases of fetal disease; murine studies have not shown evidence
of vertical transmission.[3] However, thorough placental pathology and microbiology assessment should be considered
in suspected RMSF.
Prompt and timely treatment with doxycycline significantly limits the morbidity and
mortality of RMSF in both children and adults.[1] However, tetracycline use in pregnancy has been limited due to classification as
category D and concern for teeth and bone malformation in the fetus and hepatotoxicity
and pancreatitis in the mother.[2]
[11] Chloramphenicol had been historically recommended for pregnant patients with RMSF
despite the risk of gray baby syndrome.[12] There is growing evidence that the use of doxycycline in pregnancy is safer than
previously thought.[11] Fetal teeth discoloration may be due to calcification occurring after delivery;
therefore, short courses of doxycycline may be acceptable.[3] Prior case studies have also demonstrated no significant sequelae to fetus or mother
with the use of doxycycline in the third trimester.[3]
[6] A recent review noted no evidence of any human teratogenicity, permanent teeth staining,
hepatotoxicity, nor permanent inhibition of bone growth effects with the use of doxycycline
in pregnant patients. Initial studies demonstrating adverse effects examined significantly
higher doses of other IV tetracyclines which are no longer used today.[11] Therefore, doxycycline should be recommended as the drug of choice for RMSF therapy
in pregnancy, particularly in severely ill patients prior to confirmatory testing.[2]
RMSF is a clinically challenging diagnosis in pregnancy given variable presentation
and other likely differential diagnoses. Many patients do not report any exposure
risk to ticks, and atypical presentations can occur as in this case. Avoidance of
tick-infested areas, proper use of clothing and repellants, and awareness of local
prevalence is important to prevent this potentially fatal infection.[2] Doxycycline use may have important implications in the setting of pregnant patients
not improving with traditional broad-spectrum antibiotics, and potentially decrease
the risk of prematurity due to iatrogenic delivery. Given the potentially dire implications
in pregnancy, providers in endemic areas should consider RMSF in the differential
of atypical presentations in pregnant women and those with a fever of unknown origin.
