 |
|
Clinical, Laboratory, and Epidemiologic Features of Murine Typhus in 97 Texas Children
Sarah F. Whiteford, MD;
Jeffery P. Taylor, MPH;
J. Stephen Dumler, MD
Arch Pediatr Adolesc Med. 2001;155:396-400.
ABSTRACT
| |
Objective To document the clinical, laboratory, and epidemiologic characteristics
of pediatric patients with murine typhus.
Design Pediatric patients were diagnosed using serologic testing, and clinical,
laboratory, and epidemiologic data were retrospectively reviewed.
Setting Of 97 patients, 77 (79%) were identified and treated as inpatients and
20 (21%) were treated as outpatients; most resided in south Texas.
Patients Between 1979 and 1996, medical records and patient-physician interviews
were available for 97 patients aged 16 years and younger with murine typhus.
Main Outcome Measures The frequency of clinical symptoms and signs, abnormal laboratory findings,
epidemiologic findings, and measures of disease severity were determined.
Results The clinical triad of fever, headache, and rash occurred in only 43
(49%) of 87 pediatric patients throughout the illness. Musculoskeletal symptoms
were experienced by 43% of patients, whereas gastrointestinal tract symptoms
(nausea, vomiting, anorexia, and diarrhea) occurred in 77%. Systemic involvement
was evident by the frequent occurrence of abnormal laboratory findings referable
to multiple organ systems, including the liver, kidney, blood, and central
nervous system.
Conclusions Pediatric infection by Rickettsia typhi usually
causes mild to moderate systemic illness. In children, the median duration
of illness was 12 days (range, 5-29 days), but severe complications were rare.
Length of illness was significantly related to the initial diagnosis, whereas
the interval to defervescence was related to therapy with a tetracycline or
chloramphenicol. Early recognition and treatment is important to prevent prolonged
morbidity.
INTRODUCTION
MURINE TYPHUS is an endemic zoonosis caused by a small obligate intracellular
bacterium called Rickettsia typhi that is transmitted
by fleas. The disease occurs worldwide, particularly in warm climates with
heavy populations of rat or opossum reservoirs and flea vectors.1, 2, 3
A dramatic reduction in reported cases of murine typhus began in the 1940s,
and the current prevalence of the disease is fewer than 100 cases per year.4, 5 Despite this fact, murine typhus is
still the second most frequently reported rickettsial infection in the United
States.4, 6 Murine typhus often
goes unrecognized and is perceived as a clinically mild disease. Texas and
regions of southern California have the highest prevalence in the United States,
but recent epidemiologic studies have stimulated concern that typhus reservoirs
and vectors are spreading.7 Although most patients
are adults, children constitute up to 75% of infections in some outbreaks.1 Despite this situation, studies with large series
of pediatric patients with murine typhus are lacking. This study retrospectively
reviews the clinical, laboratory, and epidemiologic findings in 97 pediatric
patients with murine typhus diagnosed in Texas from 1979 through 1996.
PATIENTS AND METHODS
All pediatric patients (aged  16 years) with a serologic diagnosis
of murine typhus identified between January 1, 1979, and December 31, 1996,
by the Bureau of Laboratories of the Texas Department of Health, Austin, were
included. Cases were confirmed by a 4-fold or greater increase in serum antibody
titers to typhus group antigen by indirect fluorescent antibody (IFA) testing
or by a single high titer of 128 or greater by IFA and a lower titer to spotted
fever group (Rickettsia rickettsii) antigen than
to typhus group antigen by IFA. Patient IFA titers to R
typhi greater than 2048 were recorded as 2048 because not all patients'
serum samples were titrated to an end point.
Clinical and epidemiologic data were collected from patient-physician
interviews at the time of illness or by retrospective review of medical records.
Information for all 97 hospitalized patients or outpatients was recorded on
standardized Texas Department of Health surveillance forms. Demographic features
and vector exposure were documented for all patients. Signs and symptoms were
noted as present or absent. Not all data were available for all patients;
thus, denominators used for various calculations differ. For laboratory data,
each result was classified as within or outside the reference range by standard
criteria. Antirickettsial therapy was noted, and response to treatment was
determined.
Disease severity was estimated by 3 measurements: length of febrile
illness; length of hospitalization; and semiquantitative summation of the
number of occurrences of clinical signs and symptoms or the presence of radiographic
evidence of pneumonitis, weighting neurologic signs and pneumonitis 3 times
more than other findings. Correlation analysis with linear regression was
used to determine statistically significant relations between laboratory variables
and length of illness, hospitalization, and number of clinical signs. Other
historical, clinical, epidemiologic, and laboratory variables were studied
for significant relations by correlation analyses,  2 tests,
and t tests.
RESULTS
DEMOGRAPHIC AND EPIDEMIOLOGIC FINDINGS
From 1979 through 1996, 501 murine typhus cases were reported in Texas;
97 were pediatric. The median age of pediatric patients was 8 years (range,
5 months to 16 years). There was a near equal sex distribution: 44 girls (45%)
and 53 boys (55%). Sixty-three patients (65%) were Hispanic, 30 (31%) were
white, and 1 (1%) was black. Disease onset described in 42 patients was subjectively
characterized as extremely abrupt in 14 (33%), moderately abrupt in 16 (38%),
and gradual in 12 (29%).
Flea bites were reported in 23 (34%) of 68 patients at presentation,
and flea exposure, defined as close patient-flea contact in flea-infested
areas or in areas where animal reservoirs are found (rats, cats, or opossums),
was noted for 45 (85%) of 53 patients. Of those with flea exposure, 23 (55%)
reported a flea bite; no information about flea bites was available in 3 flea-exposed
patients. Twenty-eight (52%) of 54 patients for whom month of illness was
provided were ill during the late spring and early summer months: 9 in May
(17%), 11 in June (20%), and 8 in July (15%). However, 26 patients (48%) were
ill in the remaining months, including fall and winter.
CLINICAL MANIFESTATIONS
Signs and symptoms recorded in patients during their illness are summarized
in Table 1. The typical manifestations
of rickettsial disease, fever and headache, were common (96 [100%] of 96 patients
and 67 [76%] of 88 patients, respectively) and occurred early in the illness
at the time of presentation (between days 1 and 2 of illness). A rash was
noted for 59 (63%) of 94 patients at some point during the illness (median
of 6 days after onset). The triad of fever, headache, and rash was present
in 43 (49%) of 87 patients during the illness; however, 83 (90%) of 92 patients
exhibited 2 of these signs and 11 (13%) of 87 manifested fever only. Only
14% (5/35) of patients for whom these intervals were available had this clinical
triad within the first 3 days of illness. Table 2 lists rash qualities and their frequency. In 46 patients
for whom the date of rash onset was recorded, the rash occurred a median of
6.5 days after onset of illness.
|
|
|
|
Table 1. Signs and Symptoms Recorded for Pediatric Patients Diagnosed
as Having Murine Typhus in South Texas
|
|
|
|
|
|
|
Table 2. Cutaneous Findings in 94 Pediatric Patients With Murine Typhus
|
|
|
Although fever, headache, or both were frequent at presentation, these
manifestations were reported in only 3 (27%) of 11 children aged 4 years or
younger and were much more common in older children (65 [84%] of 77 children).
Meningismus or meningeal findings were not reported despite the initial diagnosis
of meningitis in 3 patients, aged 8 and 11 years and 7 months. Cerebrospinal
fluid was not examined in any of the patients.
LABORATORY TESTS
All laboratory tests were performed variably among the patients (Table 3). Relevant findings included leukopenia
in 37% (25/68) of patients, absolute neutropenia in 68% (39/57), and absolute
lymphopenia in 51% (29/57); differential leukocyte counts showed a left shift
for 40 (77%) of 52 patients tested. Of 26 patients tested, 21 (81%) had high
erythrocyte sedimentation rates. Significant abnormalities in serum electrolyte
values in 43 patients tested included hyponatremia in 25 (58%) and hypokalemia
in 9 (21%). Liver function tests were infrequently performed but frequently
had abnormal findings. Serum urea nitrogen level was elevated in only 1 patient;
however, serum urea nitrogen–creatinine ratios were elevated ( 20)
in 6 (21%) of 29 patients tested. No evidence of renal tubular or parenchymal
disease was detected in 49 patients who underwent urinalysis.
|
|
|
|
Table 3. Laboratory Findings in Pediatric Patients With Murine Typhus*
|
|
|
The IFA titers were available for 85 patients and date of serologic
test was available for 84 patients. The first serum sample for diagnostic
confirmation was usually obtained within 3 weeks after onset of illness (median,
day 17). Figure 1 shows the mean
titers of R typhi antibodies and the cumulative percentage
of patients with serologic confirmation of infection by time after onset of
illness. Diagnostic levels of antibodies were detected by 7 days in 13 patients
(15%), by 14 days in 52 (62%), by 28 days in 72 (86%), and thereafter in all.
|
|
|
|
Mean titers of Rickettsia typhi antibodies and the percentage
of positive serum samples in 84 children with murine typhus vs time after
onset of illness. The percentage positive value associated with each interval
indicates the cumulative percentage of serum samples obtained by that interval
with a diagnostic R typhi antibody titer of 128 or greater.
|
|
|
THERAPY
Treatment data were available for 93 patients. Antibiotics were used
in 89 patients (96%). A tetracycline or chloramphenicol was used in 79 patients
(89%): 64 received a tetracycline, 13 received chloramphenicol, and 2 received
both. Ten patients received alternative antimicrobial agents.
SEVERITY
The median interval to defervescence after tetracycline therapy was
2 days (range, 0-6 days) and after chloramphenicol therapy was 2.5 days (range,
0-11 days) (P = .28). For other antimicrobial agents,
the median interval was 6 days (range, 1-14 days; P
= .003). At least 56 patients (58%) were hospitalized, and the median hospital
stay was 6 days (longest, 21 days). The duration of hospitalization was not
different for patients who received tetracycline or chloramphenicol treatment
vs other antibiotics (median, 6 days; P = .09). Length
of hospitalization was shorter only for tetracycline-treated patients vs those
taking other antibiotics (5 vs 6 days; P = .049).
Length of febrile illness was 5 to 29 days (median, 12 days). Although the
illness lasted longer in patients not treated with a tetracycline or chloramphenicol
(median, 11.0 vs 11.5 days), the difference was not significant (P = .22). However, patients with an initial diagnosis of murine typhus
had a significantly shorter illness duration (median, 8 vs 12 days; P = .002). Length of illness was not different between
patients with or without rash (median, 11 days); however, the presence of
a rash was associated with the initial diagnosis of murine typhus (P<.02, 2 test). No consistent association of laboratory
test results with all 3 measures of severity was identified.
INITIAL DIAGNOSIS
Only 32 (34%) of 95 patients were diagnosed initially as having murine
typhus. Twenty-four (25%) were diagnosed as having fever of undetermined origin.
The remainder of diagnoses reflected individual symptoms and signs typically
referable to specific organ systems. Flea bites and exposure were more frequent
in patients with murine typhus as an initial diagnosis (11 of 27 vs 12 of
41 and 26 of 27 vs 19 of 26 patients, respectively), but the differences were
not statistically significant (P = .15 and .07, respectively, 2 test). Rash was present in 81% of patients (25/31) initially diagnosed
as having murine typhus and in only 34 (56%) of the 61 patients who did not
receive this initial diagnosis (P<.02, 2 test), and thus this finding was often used as the primary basis for
clinical diagnosis. When the triad of fever, headache, and rash was detected
at any time during the illness, the initial diagnosis of murine typhus was
rendered 47% (20 of 43 patients) of the time, and the triad was more frequent
in patients with an initial diagnosis of murine typhus than in those with
other initial diagnoses (P = .01). The day of rash
detection apparently did not affect the initial diagnosis (5.6 days with rash
vs 5.4 days without rash; P = .43). Only 3 (9%) of
32 patients initially diagnosed as having murine typhus were aged 4 years
or younger compared with 13 (21%) of 63 with other initial diagnoses (P = .04).
COMMENT
Murine typhus, first recognized as a distinct entity in 1926, remains
a major health risk in developing nations and continues to occur with high
prevalence in parts of Texas and southern California.1
Although it is largely perceived as clinically mild, up to 10% of adults with
murine typhus require intensive care, and mortality has approached 4%.7 Most documented cases of murine typhus in the United
States are in adults; however, children can make up 75% of cases in other
areas of the world.1, 8, 9, 10, 11
Despite these data, pediatric illness with murine typhus has been particularly
neglected. The prevalence of murine typhus in southern and central Texas gave
us the opportunity to study this underrecognized pediatric infection. Aside
from the higher rate of headache reported in children aged 5 years or older,
no other specific finding was significantly associated with age. Moreover,
these data indicate that very young children are equally susceptible to R typhi infection but are not more severely affected.
This case series analyzed pediatric patients cared for in a primary
care setting, either as hospital inpatients or as outpatients, and displaying
varying illness severity. Unlike studies of adult populations, our study showed
a near equal sex distribution, possibly because of parental initiative or
equivalent exposure of children of both sexes. The demographic data indicate
that 63 patients (65%) were Hispanic, and the overall distribution approximates
the ethnic distribution of south Texas. Exposure to the vector is necessary
for inoculation of the rickettsia; however, only 23 (34%) of 68 patients reported
a flea bite, and of 42 patients (85%) for whom flea exposure was reported,
23 (55%) also reported flea bite. Thus, the lack of flea bite history should
not exclude murine typhus, but its identification can be an important clue
because it was present in 26 patients (96%) initially diagnosed with murine
typhus.
The triad of fever, headache, and rash has historically been used as
a clinical diagnostic tool in rickettsial disease, including murine typhus.4, 12, 13 Although all 3 of
these symptoms and signs were frequent in our patients, only 43 (49%) of 87
patients displayed all 3 simultaneously. These data illustrate the inconsistent
usefulness and unreliability of this triad of clinical findings in the diagnosis
of murine typhus. Also, rash appeared an average of 5 days after onset of
symptoms and was thus an unreliable early indicator of murine typhus. Macular
and maculopapular rashes were frequent; however, only 10% of patients had
petechiae (5 of 48 with rash description). Unlike findings noted in previous
studies of pediatric murine typhus, headache was frequently noted, especially
in older children.
Gastrointestinal tract symptoms were also frequent and might confound
diagnosis, leading to inappropriate surgical intervention.4
As opposed to another study,10 hepatomegaly
was noted in only 3 (4%) of 73 patients, potentially explained by the early
treatment in patients in Texas or by the presence of other infections in more
tropical settings. The presence of elevated serum activities of transaminases,
lactate dehydrogenase, and alkaline phosphatase indicate frequent but mild
liver injury possibly due to hepatocyte and biliary injury adjacent to inflamed
microvasculature. Despite this, mild hyperbilirubinemia was infrequent and
jaundice was absent. Similarly, R typhi systemic
vasculitis allows extravasation of protein and electrolytes, consistent with
the hypoproteinemia and hypoalbuminemia that were frequently observed. Direct
renal injury is not a frequent component of pediatric murine typhus.
Some simple laboratory tests might help suggest the diagnosis in the
correct clinical setting. Most patients had leukocyte counts within the reference
ranges, consistent with previous findings in adults.4, 12, 14
Because a systemic infection is usually associated with leukocytosis, the
presence of leukopenia or a normal leukocyte count in an ill child, especially
linked with a left shift, is a helpful finding. Moreover, the presence of
leukopenia or thrombocytopenia in a febrile patient also with evidence of
mild hepatic injury and flea exposure would suggest the possibility of murine
typhus or other rickettsial infection.
Complications of murine typhus are infrequent in children.10
Central nervous system manifestations were relatively few in this study, but
6 patients experienced ataxia and stupor, consistent with previous clinical
descriptions. In contrast, neurologic findings were reported in 15% to 45%
of adult patients.4, 11, 15
The reason for the discrepancy between children and adults is unclear but
perhaps might be due to early recognition and therapy in children or intrinsic
resistance of children to the infection. In addition to central nervous system
abnormalities, 10 patients had increased densities on chest radiographs. One
patient required intensive care, another had a relapse after initial therapy,
and a third had an appendectomy. There were no other significant complications
and no fatalities, although a case fatality rate of 1% to 4% is noted in adults.1 Regardless, our results indicate that a substantial
proportion of patients had severe illness characterized by a febrile interval
of 14 days or more (23%; 19 of 81 patients) or hospitalization of 7 days or
more (36%; 20 of 56 patients). Thus, murine typhus has the potential to cause
significant and prolonged morbidity in children, with undefined long-term
sequelae. When murine typhus is suspected, treatment and confirmation should
not be delayed. It is well documented that tetracycline and chloramphenicol
are effective agents to which R typhi–infected
patients respond promptly.4 In our study, the
median 2-day interval until defervescence after therapy supports the effectiveness
of this regimen for murine typhus.
Murine typhus is infrequently suspected in pediatric patients; diagnoses
other than murine typhus were initially rendered for most. Because patients
present with single-system complaints, the initial diagnosis is frequently
erroneous. However, murine typhus was the single most common initial diagnosis
probably because of physicians having a high index of suspicion in an endemic
area. Although the increased early recognition demonstrated here is encouraging
compared with previous studies,4 the data suggest
an overreliance on the detection of rash or the clinical triad of fever, headache,
and rash before diagnosis and specific therapy. This is important because
early diagnosis and treatment leads to shorter illness and hospitalization.
Nationwide, there has been increased concern about the emergence of
tickborne diseases such as Lyme disease, ehrlichioses, and Rocky Mountain
spotted fever. In Texas, the potential for reemergence and more widespread
occurrence of murine typhus might be even greater than for these tickborne
infections. From 1995 through 1998, 211 cases of murine typhus were reported
in Texas, whereas only 18 cases of Rocky Mountain spotted fever and 6 cases
of ehrlichiosis were reported.16 The potential
for reemergence of this fleaborne disease might also be present in other states.
Early recognition of emerging and reemerging infections depends on the ability
of clinicians to identify specific clinical and laboratory findings characteristic
of the disease. The findings in this study provide increased knowledge of
such manifestations in murine typhus, increasing the ability to detect and
promptly respond to this poorly recognized endemic zoonosis.
AUTHOR INFORMATION
Accepted for publication November 16, 2000.
We thank the patients' physicians and numerous hospital medical records
directors for their invaluable assistance in this investigation and David
H. Walker, MD, for his careful review and enlightening comments.
From the Department of Pathology, University of Maryland School of
Medicine, Baltimore (Dr Whiteford); the Texas Department of Health, Austin
(Mr Taylor); and the Division of Medical Microbiology, Department of Pathology,
The Johns Hopkins University School of Medicine, Baltimore (Dr Dumler).
Corresponding author and reprints: J. Stephen Dumler, MD, Division
of Medical Microbiology, Department of Pathology, The Johns Hopkins University
School of Medicine, Meyer B1-193, 600 N Wolfe St, Baltimore, MD 21287 (e-mail: sdumler{at}jhmi.edu).
REFERENCES
| |
1. Dumler JS. Murine typhus. Semin Pediatr Infect Dis. 1994;5:137-142.
2. Irons JV, Bohls SW, Thurman DC, et al. Probable role of the cat flea, Ctenocephalides felis, in transmission of murine typhus. Am J Trop Med Hyg. 1944;24:359-362.
3. Azad AF, Traub R, Sofi M, et al. Experimental murine typhus infection in the cat flea, Ctenocephalides felis (Siphonaptera: Publicidae). J Med Entomol. 1984;21:675-680.
ISI
| PUBMED
4. Dumler JS, Taylor JP, Walker DH. Clinical and laboratory features of murine typhus in south Texas, 1980
through 1987. JAMA. 1991;266:1365-1370.
FREE FULL TEXT
5. Centers for Disease Control and Prevention. Summary of notifiable diseases—United States, 1989. MMWR Morb Mortal Wkly Rep. 1990;38:47-59.
6. Burgdorfer W. Murine (flea borne) typhus. In: Hunter GW, ed. Tropical Medicine. 5th
ed. Philadelphia, Pa: WB Saunders Co; 1976:110-111.
7. Schriefer ME, Sacci JB Jr, Dumler JS, Bullen MG, Azad AF. Identification of a novel rickettsial infection in a patient diagnosed
with murine typhus. J Clin Microbiol. 1994;32:949-954.
FREE FULL TEXT
8. Duffy PE, Le Guillouzic H, Gass RF, Innis BL. Murine typhus identified as a major cause of febrile illness in a camp
for displaced Khmers in Thailand. Am J Trop Med Hyg. 1990;43:520-526.
9. Gross EM, Arbeli Y, Bearman JE, et al. Spotted fever and murine typhus in the Negev desert region of Israel,
1981. Bull World Health Organ. 1984;62:301-306.
ISI
| PUBMED
10. Silpapojakul K, Chupuppakarn S, Yuthasompob S, et al. Scrub and murine typhus in children with obscure fevers in the tropics. Pediatr Infect Dis J. 1991;10:200-203.
ISI
| PUBMED
11. Stuart BM, Pullen RI. Endemic (murine) typhus fever: clinical observations of 180 cases. Ann Intern Med. 1945;23:520-536.
FREE FULL TEXT
12. Miller ES, Beeson PB. Murine typhus fever. Medicine. 1946;25:1-15.
13. Villareal T. Endemic (murine) typhus in Corpus Christi. Tex Med. 1984;80:53-54.
14. Woodward TE. Murine typhus fever: clinical signs, symptoms, and pathophysiology. In: Walker DH, ed. Biology of Rickettsial Diseases,
I. Boca Raton, Fla: CRC Press LLC; 1988:79-92.
15. Older JJ. The epidemiology of murine typhus in Texas, 1969. JAMA. 1970;214:2011-2017.
FREE FULL TEXT
16. Texas Department of Health. Epidemiology in Texas 1998 Annual Report. Austin: Texas Dept of Health; 1999.
 CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati
What's this?
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
Rickettsia typhi Infection Presenting as Subacute Meningitis
Galanakis et al.
J Child Neurol 2002;17:156-157.
ABSTRACT
|
