• View in gallery

    Annual number of cases of rickettsial infections diagnosed in two regional hospitals in Hong Kong, 1995–2005.

  • 1

    Parola P, Paddock CD, Raoult D, 2005. Tick-borne rickettsioses around the world: emerging diseases challenging old concepts. Clin Microbiol Rev 18 :719–756.

    • Search Google Scholar
    • Export Citation
  • 2

    Jensenius M, Fournier P, Raoult D, 2004. Rickettsioses and the international traveler. Clin Infect Dis 39 :1493–1499.

  • 3

    Chapman AS, Bakken JS, Folk SM, Paddock CD, Bloch KC, Krusell A, Sexton DJ, Buckingham SC, Marshall GC, Storch GA, Dasch GA, McQuiston JH, Swerdlow DL, Dumler SJ, Nicholson WL, Walker DH, Eremeeva ME, Ohl CA, 2006. Diagnosis and management of tickborne rickettsial diseases: Rocky Mountain spotted fever, ehrlichioses, and anaplasmosis–United States. A practical guide for physicians and other health-care and public health professionals. MMWR Recomm Rep 55 :1–27.

    • Search Google Scholar
    • Export Citation
  • 4

    Kodama K, Senba T, Yamauchi H, Nomura T, Chikahira Y, 2003. Clinical study of Japanese spotted fever and its aggravating factors. J Infect Chemother 9 :83–87.

    • Search Google Scholar
    • Export Citation
  • 5

    Jang WJ, Kim JH, Choi YJ, Jung KD, Kim YG, Lee SH, Choi MS, Kim IS, Walker DH, Park KH, 2004. First serologic evidence of human spotted fever group rickettsiosis in Korea. J Clin Microbiol 42 :2310–2313.

    • Search Google Scholar
    • Export Citation
  • 6

    Freedman DO, Weld LH, Kozarsky PE, Fisk T, Robins R, von Sonnenburg F, Keystone JS, Pandey P, Cetron MS, 2006. Geo-Sentinel Surveillance Network. Spectrum of disease and relation to place of exposure among ill returned travelers. N Engl J Med 354 :119–130.

    • Search Google Scholar
    • Export Citation
  • 7

    Wilson ME, Weld LH, Boggild A, Keystone JS, Kain KC, von Sonnenburg F, Schwartz E, 2007. GeoSentinel Surveillance Network. Fever in returned travelers: results from the Geo-Sentinel Surveillance Network. Clin Infect Dis 44 :1560–1568.

    • Search Google Scholar
    • Export Citation
  • 8

    McQuiston JG, Paddock CE, Singleton J Jr, Wheeling JT, Zaki S, Childs JE, 2004. Imported spotted fever rickettsioses in United States travelers returning from Africa: a summary of cases confirmed by laboratory testing at the centers for disease control and prevention, 1999–2002. Am J Trop Med Hyg 70 :98–101.

    • Search Google Scholar
    • Export Citation
  • 9

    Centre for Health Protection, Hong Kong SAR, 2005. Epidemiology of Spotted Fever, Scrub Typhus and Murine Typhus in Hong Kong, 1995–2004. Public Health and Epidemiology Bulletin. Available at: http://www.chp.gov.hk/files/pdf/grp-PHEB-V14n1-en-20050322.pdf

  • 10

    Centre for Health Protection, Hong Kong SAR, 2005. Rickettsial Disease Associated with Outdoor Settings. Communicable Diseases Watch. Available at: http://www.chp.gov.hk/files/pdf/CDW_V2_23b.pdf

  • 11

    Walker DH, 2002. Rickettsia rickettsii: as virulent as ever. Am J Trop Med Hyg 66 :448–449.

  • 12

    Walker DH, 2006. Targeting rickettsia. N Engl J Med 354 :1418–1420.

  • 13

    Walker DH, 2007. Rickettsiae and rickettsial infections: the current state of knowledge. Clin Infect Dis 45 (Suppl 1):S39–S44.

  • 14

    Conlon PJ, Procop GW, Fowler V, Ali Eloubeildi M, Smith SR, Sexton DJ, 1996. Predictors of prognosis and risk of acute renal failure in patients with Rocky Mountain spotted fever. Am J Med 101 :621–626.

    • Search Google Scholar
    • Export Citation
  • 15

    Childs JE, Paddock CD, 2002. Passive surveillance as an instrument to identify risk factors for fatal Rocky Mountain spotted fever: is there more to learn? Am J Trop Med Hyg 66 :450–457.

    • Search Google Scholar
    • Export Citation
  • 16

    Holman RC, Paddock CD, Curns AT, Krebs JW, McQuiston JH, Childs JE, 2001. Analysis of risk factors for fatal Rocky Mountain spotted fever: evidence for superiority of tetracyclines for therapy. J Infect Dis 184 :1437–1444.

    • Search Google Scholar
    • Export Citation
  • 17

    Lui G, Lee N, Ip M, Choi KW, Tso YK, Lam E, Chau S, Lai R, Cockram CS, 2006. Cryptococcosis in apparently immunocompetent patients. QJM 99 :143–151.

    • Search Google Scholar
    • Export Citation
  • 18

    Sung JJ, Wu A, Joynt GM, Yuen KY, Lee N, Chan PK, Cockram CS, Ahuja AT, Yu LM, Wong VW, Hui DS, 2004. Severe acute respiratory syndrome: report of treatment and outcome after a major outbreak. Thorax 59 :414–420.

    • Search Google Scholar
    • Export Citation
  • 19

    Lee N, Hui D, Wu A, Chan P, Cameron P, Joynt GM, Ahuja A, Yung MY, Leung CB, To KF, Lui SF, Szeto CC, Chung S, Sung JJ, 2003. A major outbreak of severe acute respiratory syndrome in Hong Kong. N Engl J Med 248 :1986–1994.

    • Search Google Scholar
    • Export Citation
  • 20

    Weiss E, 1992. Encyclopedia of Microbiology. Volume 3. San Diego, CA: Academic Press, 585–610.

  • 21

    Centre for Health Protection, Hong Kong SAR, 2005. Laboratory Diagnosis of Spotted Fever and Typhus. Public Health and Epidemiology Bulletin. Mar 2005. Available at: http://www.chp.gov.hk/files/pdf/grp-PHEB-V14n1-en-20050322.pdf

  • 22

    Graves S, Wang L, Nack Z, Jones S, 1999. Rickettsia serosurvey in Kimberley, Western Australia. Am J Trop Med Hyg 60 :786–789.

  • 23

    Basnyat B, Belbase RH, Zimmerman MD, Woods CW, Reller LB, Murdoch DR, 2006. Clinical features of scrub typhus. Clin Infect Dis 42 :1505–1506.

    • Search Google Scholar
    • Export Citation
  • 24

    Dumler JS, Walker DH, 2005. Rocky Mountain spotted fever–changing ecology and persisting virulence. N Engl J Med 353 :551–553.

  • 25

    Archibald LK, Sexton DJ, 1995. Long-term sequelae of Rocky Mountain spotted fever. Clin Infect Dis 20 :1122–1125.

  • 26

    Thap LC, Supanaranond W, Treeprasertsuk S, Kitvatanachai S, Chinprasatsak S, Phonrat B, 2002. Septic shock secondary to scrub typhus: characteristics and complications. Southeast Asian J Trop Med Public Health 33 :780–786.

    • Search Google Scholar
    • Export Citation
  • 27

    Wang CC, Liu SF, Liu JW, Chung YH, Su MC, Lin MC, 2007. Acute respiratory distress syndrome in scrub typhus. Am J Trop Med Hyg 76 :1148–1152.

    • Search Google Scholar
    • Export Citation
  • 28

    Charoensak A, Chawalparit O, Suttinont C, Niwattayakul L, Losuwanaluk K, Silpasakorn S, Suputtamongkol Y, 2006. Scrub typhus: chest radiographic and clinical findings in 130 Thai patients. J Med Assoc Thai 89 :600–607.

    • Search Google Scholar
    • Export Citation
  • 29

    Kirkland KB, Wilkinson WE, Sexton DJ, 1995. Therapeutic delay and mortality in cases of Rocky Mountain spotted fever. Clin Infect Dis 20 :1118–1121.

    • Search Google Scholar
    • Export Citation
  • 30

    Kim DM, Won KJ, Park CY, Yu KD, Kim HS, Yang TY, Lee JH, Kim HK, Song HJ, Lee SH, Shin H, 2007. Distribution of eschars on the body of scrub typhus patients: a prospective study. Am J Trop Med Hyg 76 :806–809.

    • Search Google Scholar
    • Export Citation
  • 31

    Demma LJ, Traeger MS, Nicholson WL, Paddock CD, Blau DM, Eremeeva ME, Dasch GA, Levin ML, Singleton J Jr, Zaki SR, Cheek JE, Swerdlow DL, McQuiston JH, 2005. Rocky Mountain spotted fever from an unexpected tick vector in Arizona. N Engl J Med 535 :587–594.

    • Search Google Scholar
    • Export Citation
  • 32

    Zanetti G, Francioli P, Tagan D, Paddock CD, Zaki SR, 1998. Imported epidemic typhus. Lancet 352 :1709.

  • 33

    Cascio A, Colomba C, Antinori S, Paterson DL, Titone L, 2002. Clarithromycin versus azithromycin in the treatment of Mediterranean spotted fever in children: a randomized controlled trial. Clin Infect Dis 34 :154–158.

    • Search Google Scholar
    • Export Citation
  • 34

    Phimda K, Hoontrakul S, Suttinont C, Chareonwat S, Losuwanaluk K, Chueasuwanchai S, Chierakul W, Suwancharoen D, Silpasakorn S, Saisongkorh W, Peacock SJ, Day NP, Suputtamongkol Y, 2007. Doxycycline versus azithromycin for treatment of leptospirosis and scrub typhus. Antimicrob Agents Chemother 51 :3259–3263.

    • Search Google Scholar
    • Export Citation
  • 35

    Kim DM, Kim HL, Park CY, Yang TY, Lee JH, Yang JT, Shim SK, Lee SH, 2006. Clinical usefulness of eschar polymerase chain reaction for the diagnosis of scrub typhus: a prospective study. Clin Infect Dis 43 :1296–1300.

    • Search Google Scholar
    • Export Citation
  • 36

    Lee N, Rainer TH, Ip M, Zee B, Ng MH, Antonio GE, Chan E, Lui G, Cockram CS, Sung JJ, Hui DS, 2006. Role of laboratory variables in differentiating SARS-coronavirus from other causes of community-acquired pneumonia within the first 72 h of hospitalization. Eur J Clin Microbiol Infect Dis 25 :765–772.

    • Search Google Scholar
    • Export Citation
  • 37

    Leung GM, Rainer TH, Lau FL, Wong IO, Tong A, Wong TW, Kong JH, Hedley AJ, Lam TH; Hospital Authority SARS Collaborative Group, 2004. A clinical prediction rule for diagnosing severe acute respiratory syndrome in the emergency department. Ann Intern Med 141 :333–342.

    • Search Google Scholar
    • Export Citation
  • 38

    Watt G, Jongsakul K, Chouriyagune C, Paris R, 2003. Differentiating dengue virus infection from scrub typhus in Thai adults with fever. Am J Trop Med Hyg 68 :536–538.

    • Search Google Scholar
    • Export Citation
  • 39

    Sonthayanon P, Chierakul W, Wuthiekanun V, Blacksell SD, Pimda K, Suputtamongkol Y, Pukrittayakamee S, White NJ, Day NP, Peacock SJ, 2006. Rapid diagnosis of scrub typhus in rural Thailand using polymerase chain reaction. Am J Trop Med Hyg 75 :1099–1102.

    • Search Google Scholar
    • Export Citation
  • 40

    Singhsilarak T, Leowattana W, Looareesuwan S, Wongchotigul V, Jiang J, Richards AL, Watt G, 2005. Short report: detection of Orientia tsutsugamushi in clinical samples by quantitative real-time polymerase chain reaction. Am J Trop Med Hyg 72 :640–641.

    • Search Google Scholar
    • Export Citation
  • 41

    Watt G, Chouriyagune C, Ruangweerayud R, Watcharapichat P, Phulsuksombati D, Jongsakul K, Teja-Isavadharm P, Bhodhidatta D, Corcoran KD, Dasch GA, Strickman D, 1996. Scrub typhus infections poorly responsive to antibiotics in northern Thailand. Lancet 348 :86–89.

    • Search Google Scholar
    • Export Citation
  • 42

    Watt G, Kantipong P, Jongsakul K, Watcharapichat P, Phulsuksombati D, Strickman D, 2000. Doxycycline and rifampicin for mild scrub-typhus infections in northern Thailand: a randomised trial. Lancet 356 :1057–1061.

    • Search Google Scholar
    • Export Citation

 

 

 

 

Risk Factors Associated with Life-threatening Rickettsial Infections

View More View Less
  • 1 Department of Medicine and Therapeutics, and Department of Microbiology, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, People’s Republic of China; Department of Medicine and Geriatrics, and Department of Pathology, Princess Margaret Hospital, Hong Kong, People’s Republic of China

We retrospectively analyzed 92 cases of severe rickettsial infections in patients (median age = 49 years, 57% male, 37.0% with scrub typhus) in Hong Kong. Immunofluorescence assay was used for diagnostic confirmation. Identification of ≥ 1 diagnostic sign (exposure history, rash, or eschar) was possible in 94.6% of the cases. Multivariate analysis suggested that pulmonary infiltrates (odds ratio [OR] = 25.2, 95% confidence interval [CI] = 3.9–160.9, P = 0.001) and leukocytosis (OR = 1.3, 95% CI = 1.0–1.5 per unit increase, P = 0.033) were independent predictors of admission to an intensive care unit (14.1%). Delayed administration of doxycycline was independently associated with major organ dysfunction (23.9%; oxygen desaturation, renal failure, severe jaundice, encephalopathy, cardiac failure) (OR = 1.2, 95% CI = 1.0–1.5 per day delay, P = 0.046; adjusted for age and rickettsia biogroup) and prolonged hospitalization > 10 days (25%) (OR = 1.4, 95% CI = 1.1–1.9 per day delay, P = 0.014). Treatment with fluoroquinolone/clarithromycin did not correlate with clinical outcomes (P > 0.05). Early empirical doxycycline therapy should be considered if clinico-epidemiologic signs of rickettsial infections are present.

INTRODUCTION

Rickettsioses are acute febrile, arthropod-borne diseases caused by obligate intracellular bacteria, and are classified into three major biogroups: scrub typhus (caused by Orientia tsutsugamushi, which is widely endemic in Asia), spotted fever (e.g., caused by Rickettsia rickettsii, R. conorii, R. africae, and R. japonica), and murine/epidemic typhus (caused by Rickettsia typhi and R. prowazekii).15 These infections are endemic, some of them are emerging, and they are important causes of fever in certain populations and travelers.68 In Hong Kong, more than 200 cases of spotted fever, scrub typhus, and murine typhus (no epidemic typhus) have been reported in the past decade, with an increasing trend observed in recent years.9,10 The disease spectrum of rickettsioses is wide. In some patients, the disease is mild; however, other patients develop serious complications and fatalities are not uncommonly reported (up to 35% with scrub typhus).13 However, risk factors associated with these life-threatening rickettsial infections remain unclear.24,1116 In this study, potential risk factors were examined in a large cohort of hospitalized patients. These included baseline characteristics, rickettsia biogroup, clinical and radiological variables, and delay in specific treatment. Understanding such factors may improve management and outcomes of these patients. We herein report that scrub typhus, pulmonary infiltration, leukocytosis, and delayed doxycycline therapy are likely important risk factors for life-threatening rickettsial infections.

PATIENTS AND METHODS

A retrospective analysis was performed. All adult (age > 16 years) cases of rickettsial infections diagnosed and managed in two major acute general hospitals (each with > 1,200 beds and serving urban and sub-urban areas) in Hong Kong over a 10-year period (1995–2005) were studied. Cases were identified through the hospitals’ computerized clinical management system (searching ICD-9 codes for rickettsioses, spotted fevers, and typhus), and the laboratory information system of the microbiology laboratories.17 Cases were included for analysis if the diagnosis was confirmed by immunofluorescence assay (IFA), or if supportive serologic findings were available (see below). In the latter situation, cases should also include a febrile illness characterized by eschar (tache noir) formation and/or skin rash, and with no alternative diagnosis.3,9,10,16 Seven cases were excluded because of no serologic support.

All patients’ case notes and laboratory and radiology records were reviewed. Clinical, routine laboratory, microbiologic, and radiologic variables were compared among patients with or without serious, life-threatening illnesses,4,14,16 including intensive care unit (ICU) admission (for the management of septic shock, severe respiratory failure, acute respiratory distress syndrome, and initiation of dialysis)18,19; development of one or more major organ dysfunction, including oxygen desaturation, acute renal failure, severe jaundice, encephalopathy, and cardiac failure (± ICU admission)13; and prolonged hospitalization beyond 10 days (i.e., length of stay > 75th percentile in our cohort) as a result of persistent symptoms.3

Univariate associations were examined using a chi-square test, Fisher’s exact (two-tailed) test, or an unpaired t-test as appropriate. Variables with a P value < 0.10 in univariate analyses and age16 were entered into the multivariate models as covariates. Stepwise backward logistic regression was performed to identify independent variables associated with life-threatening diseases. A P value less than 0.05 was considered statistically significant. All probabilities were two-tailed. Statistical analysis was performed with the SPSS version 13.0 software (SPSS Inc., Chicago, IL).

Serologic testing for rickettsial infection.

Paired acute-phase and convalescent-phase serum samples collected two weeks apart (median = 13 days, interquartile range [IQR] = 9–15 days) were tested using the Weil-Felix test20 and an IFA for specific IgM and IgG antibodies against rickettsial antigens. The tests included antigens from species in the three major bio-groups (spotted fever group = R. Australis, R. honei, and R. conori; typhus group = R. prowazekii and R. typhi; scrub typhus group = O. tsutsugamushi strains). The IFA is generally regarded as the standard serologic test for rickettsial infections because of its higher sensitivity and specificity; and identification of the pathogen is possible down to the biogroup level.1,21,22 A 1 in 10 dilution of the patients’ sera were screened for positivity by IFA according to instructions from the manufacturer (Gilliam and Litchfield; Australian Rickettsial Reference Laboratory Foundation Ltd., Geelong, Victoria, Australia).22 Serial dilutions were repeated with endpoint titers recorded as the reciprocal of the last dilution exhibiting specific fluorescence. A ≥ 4-fold increase in titer or a single high antibody titer ≥ 640 to either of the spotted fever group, typhus group, or scrub typhus antigens, or a positive IgM result for scrub typhus antigens were considered indicative of current rickettsial infection.22 A reactive/positive IFA result that did not satisfy these criteria, e.g., titer ≥ 320, and/or 4-fold increase in titer by the Weil-Felix test, were regarded as supportive evidence for current rickettsial infection.

RESULTS

A total of 92 cases were studied. These cases had a median age of 49 years (IQR = 39–60 years); 56.5% were male, 6.5% had co-existing medical conditions (e.g., diabetes), 93.5% were ethnic Chinese, and 83.7% were local cases (the remainder were mostly from Guangdong province, China). During the 10-year study period, an increasing number of cases were identified each year (Figure 1), and a seasonal pattern (80.4% from June through November) was observed.9,10 Reactive IFA results were documented in 78.3% of the cases. Sixty-three (68.5%) cases fulfilled the serologic confirmation criteria (4-fold increase in titer = 58.7%, single high titer ≥ 640 = 41.3%), and 29 (31.5%) cases had supportive serologic findings (reactive IFA, n = 9; 4-fold increase in titer in Weil-Felix test, n = 20). Rickettsia biogroup as indicated by serologic results showed that 36.9% of the cases were scrub typhus, 46.7% were either spotted fever or murine typhus, and 16.3% were inconclusive (Table 1).

Clinico-epidemiologic findings were analyzed (Table 1). Possible exposure history, rash, and eschar were documented in 65.2%, 65.2%, and 52.2% of the cases respectively; and overall 87 (94.6%) cases had 1 or more of these findings reported. Abnormal liver function (> 90%) and moderate thrombocytopenia (> 60%) were commonly observed. Patients confirmed with scrub typhus were less likely to develop a rash (38.2% versus 81.4%; P < 0.001), but more likely to have an eschar (61.8% versus 40.5%; P = 0.065). Pulmonary infiltrates were also more common in patients with scrub typhus (35.3% versus 7.0%; P = 0.002). Thirteen (14.1%) patients required intensive care; 22 (23.9%) patients developed one or more major organ dysfunction, including oxygen desaturation (n = 11), acute renal failure (n = 6), severe jaundice (n = 10), encephalopathy (n = 10), and cardiac failure (n = 2). No deaths were observed in this cohort.

Initial misdiagnoses were common (90%) and included drug allergy (7.3%), hepatitis/biliary sepsis (19.5%), meningitis/pneumonia/urosepsis (26.8%), viral illness (22.0%), typhoid (4.9%), and unspecified febrile illness (19.5%). Abnormal urinalyses were observed in 64%, and an ultrasound scan to investigate for hepatic/biliary pathology was ordered by clinicians in 57.6%. Overall, 44 (47.8%) patients had received either a macrolide (clarithromycin, n = 16) and/or fluoroquinolones (levofloxacin or ciprofloxacin, n = 35) prior to doxycycline therapy. Doxycycline (n = 79, 85.9%) was initiated when the diagnosis of rickettsial infection was eventually suspected (median delay after admission 3 days, IQR = 1–4 days) in patients with on-going illness. The median time for defervescence after doxycycline initiation was 2 days (IQR = 1–3 days). Chloramphenicol was not used in any of the cases.

Factors associated with life-threatening rickettsial diseases were studied. Univariate analyses (Table 2) showed that scrub typhus (80%), pulmonary infiltrates on chest radiographs, increased plasma urea concentrations, and leukocytosis were associated with ICU admission. Secondary bacterial infection was not evident in any of these cases. Scrub typhus (P = 0.034) and failure to initiate doxycycline within the first three days of being hospitalized (P = 0.033) were associated with development of major organ dysfunction. Scrub typhus was diagnosed in 60% of these patients. Older age (P = 0.002), major organ dysfunction (P < 0.001), and failure to initiate doxycycline in the first 3 days of hospitalization (P = 0.040) were associated with prolonged hospitalization > 10 days. Other clinical variables were not found to correlate with outcomes. The use of a fluoroquinolone/clarithromycin on hospitalization was not associated with a better clinical outcome in this analysis (29.5% versus 19.1% in patients with or without major organ dysfunction, respectively; P = 0.247).

Multivariate analyses were performed to identify independent factors associated with adverse outcomes (Table 3). Pulmonary infiltrates (OR = 25.2, 95% CI = 3.9–160.9, P = 0.001) and leukocytosis (OR = 1.3, 95% CI = 1.0–1.5 per unit increase, P = 0.033) were independent predictors for ICU admission (adjusted for rickettsia biogroup and baseline characteristics). Delayed doxycycline administration after hospitalization was associated with major organ dysfunction (OR = 1.2, 95% CI = 1.0–1.5 per day delay, P = 0.046, adjusted for age and rickettsia biogroup) and prolonged hospitalization > 10 days (OR = 1.4, 95% CI = 1.1–1.9 per day delay, P = 0.014, adjusted for age, organ dysfunction, and rickettsia biogroup). Even when cases with supportive serologic findings were excluded from the analyses, these risk factors remained significant in the final multivariate models (for ICU admission: pulmonary infiltrates, OR = 23.0, P = 0.001; leukocytosis, OR = 1.25, P = 0.037; for major organ dysfunction: delayed doxycycline administration, OR = 1.28, P = 0.021).

All ICU patients received doxycycline treatment, although a fluoroquinolone or clarithromycin was initially given to 10 (77%) of 13 patients before rickettsial infection was suspected. Therapy was then switched to doxycyline at a median of 3 (IQR = 0.5–7.5) days later to treat on-going illness. Computed tomographic scans of the thorax and abdomen were performed in two patients with scrub typhus and showed lower lobe ground-glass changes, pleural effusions, ascites, hepatosplenomegaly, and dilated bowel loops or diffuse bowel wall thickening (patients had diarrhea). In another case, biopsy of a regional lymph node (proximal to the eschar) showed necrotizing lymphadenitis with areas of infarct. Large numbers of plasma cells and histiocytes were present. No organism was seen with routine staining.

DISCUSSION

Life-threatening diseases may develop in a significant proportion of patients with rickettsioses. Associated risk factors include scrub typhus, pulmonary infiltration, high leukocyte count, and delayed doxycycline therapy. Empirical therapy should therefore be considered early if clinico-epidemiologic signs (exposure history, eschar, or rash) are present.

These findings are consistent with earlier observations that scrub typhus is a severe disease (it constituted most of the life-threatening infections in this study). Pneumonitis, meningitis, multiorgan failure, and disseminated intravascular coagulation (DIC) can occur if untreated and the reported fatality rate is 1–35%.2,9,10,23 Clinical manifestations are believed to be the result of generalized vasculitis and increased microvascular permeability.2,3,1214 The spotted fever group covers a wide disease spectrum and certain species in this biogroup, e.g., R. rickettsii, can cause acute renal failure, hepatitis, acute respiratory distress syndrome, meningitis, DIC, and death if patients are not treated.14,9,10,14,24,25 In our locality, R. japonica has been identified as the most prevalent rickettsial species causing spotted fever,4,5,21 and can result in life-threatening illness.4 Findings of leukocytosis and pulmonary infiltrates being independent risk factors for life-threatening rickettsial diseases are supported by the results of recent studies. Moderate leukocytosis has been observed in serious scrub typhus26,27 and spotted fever infections,4 and has been shown to correlate with the overall systemic inflammatory responses (e.g., C-reactive protein levels and cytokines levels).4 Pulmonary infiltration is common (> 40%) in scrub typhus and has been associated with development of multi-organ failure.2628 Clinicians should be alert to the risk of deterioration if these factors are identified at presentation.

We demonstrated by multivariate analyses that delayed doxycycline therapy is independently associated with increased risk of major organ dysfunction and protracted illness (prolonged hospitalization) after adjustment for rickettsia biogroup and baseline characteristics.16 The risk increases by at least 20% with a day of delay in doxycycline treatment after presentation (the median delay in our cohort was 3 days). This finding is supported by earlier (univariate analysis) observations that late treatment of rickettsial infection ≥ 5 days from disease onset is associated with higher risk of progressive clinical deterioration and death.4,11,12,14,16,27,29 The delay in treatment may be related to atypical presentations of rickettsial infections such as pneumonitis, hepatitis, urinary abnormalities, or absence of a rash (Table 2),29 and because early laboratory diagnosis is often difficult. Serologic response is usually undetectable during early illness, and molecular testing by polymerase chain reaction (PCR) is generally unavailable.13 Given the high risk of delaying therapy, empirical doxycycline should be initiated early in a febrile patient when clinico-epidemiologic diagnostic signs for rickettsial infection are present.

Our data suggest that a relevant exposure history, rash, or an eschar could be identified in up to 95% of the cases (Table 1).2,4,30 Other important clinical features include abnormal liver function (> 90%), headache (> 65%), and moderate thrombocytopenia (> 60%).23 Epidemiologic assessment should include recent history of travel, occupational or recreational activities, and history of arthropod bites (including dog ticks).3,610,31 Rickettsial infection is now regarded as one of the most common causes of fever in returned travelers in addition to malaria, typhoid, and dengue.6,7 This is possibly related to climate/ecologic changes, increased international travel, ecotourism, and increased awareness of the disease.2

Our analysis was limited by its retrospective nature, and by the fact that only hospitalized patients were included.9,10 Mild or fatal cases not serologically tested may have been missed.3,16 Also, our analysis may be confounded by the use of fluoroquinolone (majority) or macrolide (clarithromycin, not azithromycin) in some patients too as treatment for infection with atypical pathogens before doxycycline administration. The efficacy of these antibiotics cannot be directly addressed in this study, but there was no apparent association between the use of these regimens and clinical outcomes. Doxycycline is generally considered as the drug of choice for all rickettsial infections.13 Fluoroquinolones may fail clinically despite showing good in vitro activity,2,32 and a slower response with macrolides has been observed, although clinical efficacy demonstrated for newer macrolides (e.g., azithromycin) deserves further investigation.1,11,33,34 We observed that some patients had persistent symptoms despite a few days of fluroquinolone/macrolide treatment, which resolved promptly after treatment with doxycycline.

Another limitation is the diagnosis and differentiation of the rickettsia biogroup based on serologic analysis alone rather than PCR or culture (not available in our unit).1,21,35 Even with IFA (although more sensitive and specific than the Weil-Felix test), definitive diagnosis is sometimes difficult because of delayed antibody responses or blunted responses post-treatment.9,10,21,22 Thus, patients with a typical, otherwise unexplained, illness and supportive serologic findings were included.16 In addition, although the serologic response to scrub typhus shows little cross-reactivity with the other biogroups, significant cross-reactivity was not uncommonly observed between the spotted fever and murine typhus biogroups,1,9,10,21,22 which makes their reliable differentiation difficult (although spotted fever seems twice more common, which is consistent with local surveillance data).9 Thus, we had only compared scrub typhus against spotted fever and murine typhus. Future clinical studies on rickettsial infections should include development of a diagnostic algorithm (e.g., using a clinical prediction rule approach on the basis of epidemiologic, clinical, and laboratory variables),3,3638 application of molecular techniques to assist diagnosis and speciation (e.g., PCR),13,35,39,40 prospective randomized trials for newer therapies/antimicrobial drugs,1,34,41,42 and studies on antimicrobial drug resistance.11,12,34,41,42

In conclusion, scrub typhus, pulmonary infiltration, leukocytosis, and delayed doxycycline therapy are risk factors for life-threatening rickettsial infections. Early empirical doxycycline therapy should be considered at presentation if clinico-epidemiologic signs for rickettsial infections are present.

Table 1

Clinical descriptions of 92 hospitalized patients with rickettsial infections*

Clinical observations%Clinical observations%
* Possible exposure = construction site work in sub-urban areas, patrolling the country, field trips, hiking, and camping.3 ICU = intensive care unit; LOS = median hospital length of stay; USG = ultrasound scan; LFT = liver function test; ALT = alanine aminotransferase; ULN = upper limit of normal; RBC = red blood cell; WBC = white blood cell. Abnormal LFT results23 were changes in bilirubin, ALT, or alkaline phosphatase above their ULN. Results for urinary RBCs, protein, and WBCs were from multi-stix results of 80 patients.14,26 Diagnosis of scrub typhus was based on immunofluorescence assay results (n = 34); the remaining cases were identified as spotted fever or murine typhus (n = 43) or inconclusive (n = 15, because of incomplete data or cross-reactivity) on the basis of serologic findings. Major organ dysfunction were oxygen desaturation (n = 11), acute renal failure (n = 6), severe jaundice (n = 10), encephalopathy (n = 10), and cardiac failure (n = 2). Median LOS = 8 days (interquartile range = 6–11 days). Neurologic manifestations9,10,23 were meningitis (with abnormal cerebrospinal fluid findings, n = 4), polyneuropathy (n = 3), and hearing impairment (n = 2).
Possible exposure65.2Scrub typhus37.0
Recalling insect bite19.6Death0.0
Fever98.9Major organ dysfunctions23.9
Headache65.2ICU admission14.1
Myalgia40.2Prolonged LOS > 10 days26.1
Rash65.2Correct initial diagnosis10.9
Eschar (site of eschar: trunk)52.2 (75.0)Doxycycline therapy (within 3 days of admission)85.9 (55.4)
Lymphadenopathy22.8Ever received macrolide17.4
Hepatomegaly13.0Ever received fluoroquinolone38.0
Splenomegaly15.2USG of liver performed57.6
Abnormal LFT result92.4Lumbar puncture12.0
Peak ALT > 3 times ULN42.4Urine RBC ≥ 1+39.1
Peak bilirubin > 2 times ULN15.2Urine protein ≥ 1+50.0
Platelet count < 140 × 109/L60.9Urine WBC ≥ 1+7.6
WBC > 11 × 109/L17.4
Urea > 9 mmol/L15.2
Table 2

Comparisons of patients with life-threatening rickettsiosis who required intensive care support (n = 13) with those who survived without intensive care (n = 79)*

CharacteristicsICU (n = 13)Non-ICU (n = 79)P
* ICU = intensive care unit; NS = not significant; possible exposure = construction site work in sub-urban areas, patrolling the country, field trips, hiking, and camping3; pulmonary infiltrates = radiographic infiltrates as reported by attending physicians; oxygen desaturation = requirement for supplemental oxygen to maintain oxygen saturation > 95%; ARDS = acute respiratory distress syndrome; hypotension = systolic blood pressure < 90 mm of Hg; Acute renal failure = serum creatinine > 2 mg/dL (180 μmol/L); severe jaundice = bilirubin > 60 μmol/L (> 3 times the upper limit of normal); encephalopathy = neurologic symptoms reported by the attending physician, e.g., confusion, mentally obtund; INR = international normalized ratio; APTT = activated partial thromboplastin time. Clotting profiles and C-reactive protein levels were not measured for all patients. All ICU patients received doxycycline treatment; in three patients, intravenous minocycline was first initiated and switched to doxycyline when eating was resumed. A delay in therapy after hospitalization was counted for minocycline.
† Laboratory findings on presentation.
Clinical (%)
    Age, years, mean ± SD52.2 ± 15.648.3 ± 14.1NS
    Male sex61.555.7NS
    Scrub typhus80.038.80.019
    Possible exposure53.831.6NS
    Fever10098.7NS
    Headache53.867.1NS
    Rash46.268.4NS
    Eschar61.551.3NS
    Hepatosplenomegaly23.114.1NS
    Pulmonary infiltrates76.910.1< 0.001
    Oxygen desaturation69.22.5< 0.001
    ARDS53.80< 0.001
    Hypotension66.71.3< 0.001
    Acute renal failure30.82.50.003
    Severe jaundice38.56.30.004
    Encephalopathy38.56.40.005
    Cardiac failure7.71.3NS
    Delay in doxycycline therapy, days, mean ± SD4.4 ± 3.92.9 ± 2.5NS
Laboratory† (mean ± SD)
    Sodium (mmol/L)130.4 ± 5.4132.9 ± 3.90.050
    Urea (mmol/L)9.6 ± 6.35.2 ± 3.5< 0.001
    Creatinine (μmol/L)127.4 ± 88.091.5 ± 25.8NS
    Bilirubin (μmol/L)65.7 ± 105.218.9 ± 23.6NS
    Alanine aminotransferase (IU/L)136.6 ± 90.4145.5 ± 124.9NS
    Alkaline phosphatase (IU/L)221.2 ± 134.8172.1 ± 117.5NS
    White blood cell count × 109/L11.8 ± 6.37.3 ± 3.50.026
    Platelet count × 109/L102.9 ± 42.1129.5 ± 67.3NS
    INR1.2 ± 0.11.1 ± 0.10.006
    APTT (sec)45.7 ± 12.136.4 ± 7.90.019
    C-reactive protein (mg/dL)140.2 ± 133.886.0 ± 62.2NS
Table 3

Factors associated with life-threatening illnesses in hospitalized patients with rickettsial infections*

Clinical outcomesVariables in the final multivariate modelOdds ratio (95% confidence interval)P
* ICU = intensive care unit.
ICU admissionPulmonary infiltrates25.2 (3.9–160.9)0.001
White blood cell count (per unit)1.3 (1.0–1.5)0.033
Major organ dysfunctionDelayed doxycycline (per day)1.2 (1.0–1.5)0.046
Scrub typhus3.3 (1.0–10.7)0.051
Prolonged hospitalization > 10 daysAge (per year)1.1 (1.0–1.1)0.058
Major organ dysfunction10.0 (2.3–43.9)0.002
Delayed doxycycline (per day)1.4 (1.1–1.9)0.014
Figure 1.
Figure 1.

Annual number of cases of rickettsial infections diagnosed in two regional hospitals in Hong Kong, 1995–2005.

Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 78, 6; 10.4269/ajtmh.2008.78.973

*

Address correspondence to Nelson Lee, Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, Special Administrative Region, People’s Republic of China. E-mail: leelsn@cuhk.edu.hk

Authors’ addresses: Nelson Lee, Bonnie Wong, Grace Lui, Kin Wing Choi, Jenny Ho, Yin Yan Chan, and Clive S. Cockram, Department of Medicine and Therapeutics, Prince of Wales Hospital, 9/F Clinical Science Building, The Chinese University of Hong Kong, Hong Kong, Special Administrative Region, People’s Republic of China, E-mail: leelsn@cuhk.edu.hk. Margaret Ip and Rebecca Lam, Department of Microbiology, Prince of Wales Hospital, The Chinese University of Hong Kong; Hong Kong, Special Administrative Region, People’s Republic of China. Owen Tak Yin Tsang, Jak Yiu Lai, and Sik To Lai, Department of Medicine and Geriatrics, Princess Margaret Hospital, Hong Kong, Special Administrative Region, People’s Republic of China. Tak Keung Ng, Department of Pathology, Princess Margaret Hospital, Hong Kong, Special Administrative Region, People’s Republic of China.

REFERENCES

  • 1

    Parola P, Paddock CD, Raoult D, 2005. Tick-borne rickettsioses around the world: emerging diseases challenging old concepts. Clin Microbiol Rev 18 :719–756.

    • Search Google Scholar
    • Export Citation
  • 2

    Jensenius M, Fournier P, Raoult D, 2004. Rickettsioses and the international traveler. Clin Infect Dis 39 :1493–1499.

  • 3

    Chapman AS, Bakken JS, Folk SM, Paddock CD, Bloch KC, Krusell A, Sexton DJ, Buckingham SC, Marshall GC, Storch GA, Dasch GA, McQuiston JH, Swerdlow DL, Dumler SJ, Nicholson WL, Walker DH, Eremeeva ME, Ohl CA, 2006. Diagnosis and management of tickborne rickettsial diseases: Rocky Mountain spotted fever, ehrlichioses, and anaplasmosis–United States. A practical guide for physicians and other health-care and public health professionals. MMWR Recomm Rep 55 :1–27.

    • Search Google Scholar
    • Export Citation
  • 4

    Kodama K, Senba T, Yamauchi H, Nomura T, Chikahira Y, 2003. Clinical study of Japanese spotted fever and its aggravating factors. J Infect Chemother 9 :83–87.

    • Search Google Scholar
    • Export Citation
  • 5

    Jang WJ, Kim JH, Choi YJ, Jung KD, Kim YG, Lee SH, Choi MS, Kim IS, Walker DH, Park KH, 2004. First serologic evidence of human spotted fever group rickettsiosis in Korea. J Clin Microbiol 42 :2310–2313.

    • Search Google Scholar
    • Export Citation
  • 6

    Freedman DO, Weld LH, Kozarsky PE, Fisk T, Robins R, von Sonnenburg F, Keystone JS, Pandey P, Cetron MS, 2006. Geo-Sentinel Surveillance Network. Spectrum of disease and relation to place of exposure among ill returned travelers. N Engl J Med 354 :119–130.

    • Search Google Scholar
    • Export Citation
  • 7

    Wilson ME, Weld LH, Boggild A, Keystone JS, Kain KC, von Sonnenburg F, Schwartz E, 2007. GeoSentinel Surveillance Network. Fever in returned travelers: results from the Geo-Sentinel Surveillance Network. Clin Infect Dis 44 :1560–1568.

    • Search Google Scholar
    • Export Citation
  • 8

    McQuiston JG, Paddock CE, Singleton J Jr, Wheeling JT, Zaki S, Childs JE, 2004. Imported spotted fever rickettsioses in United States travelers returning from Africa: a summary of cases confirmed by laboratory testing at the centers for disease control and prevention, 1999–2002. Am J Trop Med Hyg 70 :98–101.

    • Search Google Scholar
    • Export Citation
  • 9

    Centre for Health Protection, Hong Kong SAR, 2005. Epidemiology of Spotted Fever, Scrub Typhus and Murine Typhus in Hong Kong, 1995–2004. Public Health and Epidemiology Bulletin. Available at: http://www.chp.gov.hk/files/pdf/grp-PHEB-V14n1-en-20050322.pdf

  • 10

    Centre for Health Protection, Hong Kong SAR, 2005. Rickettsial Disease Associated with Outdoor Settings. Communicable Diseases Watch. Available at: http://www.chp.gov.hk/files/pdf/CDW_V2_23b.pdf

  • 11

    Walker DH, 2002. Rickettsia rickettsii: as virulent as ever. Am J Trop Med Hyg 66 :448–449.

  • 12

    Walker DH, 2006. Targeting rickettsia. N Engl J Med 354 :1418–1420.

  • 13

    Walker DH, 2007. Rickettsiae and rickettsial infections: the current state of knowledge. Clin Infect Dis 45 (Suppl 1):S39–S44.

  • 14

    Conlon PJ, Procop GW, Fowler V, Ali Eloubeildi M, Smith SR, Sexton DJ, 1996. Predictors of prognosis and risk of acute renal failure in patients with Rocky Mountain spotted fever. Am J Med 101 :621–626.

    • Search Google Scholar
    • Export Citation
  • 15

    Childs JE, Paddock CD, 2002. Passive surveillance as an instrument to identify risk factors for fatal Rocky Mountain spotted fever: is there more to learn? Am J Trop Med Hyg 66 :450–457.

    • Search Google Scholar
    • Export Citation
  • 16

    Holman RC, Paddock CD, Curns AT, Krebs JW, McQuiston JH, Childs JE, 2001. Analysis of risk factors for fatal Rocky Mountain spotted fever: evidence for superiority of tetracyclines for therapy. J Infect Dis 184 :1437–1444.

    • Search Google Scholar
    • Export Citation
  • 17

    Lui G, Lee N, Ip M, Choi KW, Tso YK, Lam E, Chau S, Lai R, Cockram CS, 2006. Cryptococcosis in apparently immunocompetent patients. QJM 99 :143–151.

    • Search Google Scholar
    • Export Citation
  • 18

    Sung JJ, Wu A, Joynt GM, Yuen KY, Lee N, Chan PK, Cockram CS, Ahuja AT, Yu LM, Wong VW, Hui DS, 2004. Severe acute respiratory syndrome: report of treatment and outcome after a major outbreak. Thorax 59 :414–420.

    • Search Google Scholar
    • Export Citation
  • 19

    Lee N, Hui D, Wu A, Chan P, Cameron P, Joynt GM, Ahuja A, Yung MY, Leung CB, To KF, Lui SF, Szeto CC, Chung S, Sung JJ, 2003. A major outbreak of severe acute respiratory syndrome in Hong Kong. N Engl J Med 248 :1986–1994.

    • Search Google Scholar
    • Export Citation
  • 20

    Weiss E, 1992. Encyclopedia of Microbiology. Volume 3. San Diego, CA: Academic Press, 585–610.

  • 21

    Centre for Health Protection, Hong Kong SAR, 2005. Laboratory Diagnosis of Spotted Fever and Typhus. Public Health and Epidemiology Bulletin. Mar 2005. Available at: http://www.chp.gov.hk/files/pdf/grp-PHEB-V14n1-en-20050322.pdf

  • 22

    Graves S, Wang L, Nack Z, Jones S, 1999. Rickettsia serosurvey in Kimberley, Western Australia. Am J Trop Med Hyg 60 :786–789.

  • 23

    Basnyat B, Belbase RH, Zimmerman MD, Woods CW, Reller LB, Murdoch DR, 2006. Clinical features of scrub typhus. Clin Infect Dis 42 :1505–1506.

    • Search Google Scholar
    • Export Citation
  • 24

    Dumler JS, Walker DH, 2005. Rocky Mountain spotted fever–changing ecology and persisting virulence. N Engl J Med 353 :551–553.

  • 25

    Archibald LK, Sexton DJ, 1995. Long-term sequelae of Rocky Mountain spotted fever. Clin Infect Dis 20 :1122–1125.

  • 26

    Thap LC, Supanaranond W, Treeprasertsuk S, Kitvatanachai S, Chinprasatsak S, Phonrat B, 2002. Septic shock secondary to scrub typhus: characteristics and complications. Southeast Asian J Trop Med Public Health 33 :780–786.

    • Search Google Scholar
    • Export Citation
  • 27

    Wang CC, Liu SF, Liu JW, Chung YH, Su MC, Lin MC, 2007. Acute respiratory distress syndrome in scrub typhus. Am J Trop Med Hyg 76 :1148–1152.

    • Search Google Scholar
    • Export Citation
  • 28

    Charoensak A, Chawalparit O, Suttinont C, Niwattayakul L, Losuwanaluk K, Silpasakorn S, Suputtamongkol Y, 2006. Scrub typhus: chest radiographic and clinical findings in 130 Thai patients. J Med Assoc Thai 89 :600–607.

    • Search Google Scholar
    • Export Citation
  • 29

    Kirkland KB, Wilkinson WE, Sexton DJ, 1995. Therapeutic delay and mortality in cases of Rocky Mountain spotted fever. Clin Infect Dis 20 :1118–1121.

    • Search Google Scholar
    • Export Citation
  • 30

    Kim DM, Won KJ, Park CY, Yu KD, Kim HS, Yang TY, Lee JH, Kim HK, Song HJ, Lee SH, Shin H, 2007. Distribution of eschars on the body of scrub typhus patients: a prospective study. Am J Trop Med Hyg 76 :806–809.

    • Search Google Scholar
    • Export Citation
  • 31

    Demma LJ, Traeger MS, Nicholson WL, Paddock CD, Blau DM, Eremeeva ME, Dasch GA, Levin ML, Singleton J Jr, Zaki SR, Cheek JE, Swerdlow DL, McQuiston JH, 2005. Rocky Mountain spotted fever from an unexpected tick vector in Arizona. N Engl J Med 535 :587–594.

    • Search Google Scholar
    • Export Citation
  • 32

    Zanetti G, Francioli P, Tagan D, Paddock CD, Zaki SR, 1998. Imported epidemic typhus. Lancet 352 :1709.

  • 33

    Cascio A, Colomba C, Antinori S, Paterson DL, Titone L, 2002. Clarithromycin versus azithromycin in the treatment of Mediterranean spotted fever in children: a randomized controlled trial. Clin Infect Dis 34 :154–158.

    • Search Google Scholar
    • Export Citation
  • 34

    Phimda K, Hoontrakul S, Suttinont C, Chareonwat S, Losuwanaluk K, Chueasuwanchai S, Chierakul W, Suwancharoen D, Silpasakorn S, Saisongkorh W, Peacock SJ, Day NP, Suputtamongkol Y, 2007. Doxycycline versus azithromycin for treatment of leptospirosis and scrub typhus. Antimicrob Agents Chemother 51 :3259–3263.

    • Search Google Scholar
    • Export Citation
  • 35

    Kim DM, Kim HL, Park CY, Yang TY, Lee JH, Yang JT, Shim SK, Lee SH, 2006. Clinical usefulness of eschar polymerase chain reaction for the diagnosis of scrub typhus: a prospective study. Clin Infect Dis 43 :1296–1300.

    • Search Google Scholar
    • Export Citation
  • 36

    Lee N, Rainer TH, Ip M, Zee B, Ng MH, Antonio GE, Chan E, Lui G, Cockram CS, Sung JJ, Hui DS, 2006. Role of laboratory variables in differentiating SARS-coronavirus from other causes of community-acquired pneumonia within the first 72 h of hospitalization. Eur J Clin Microbiol Infect Dis 25 :765–772.

    • Search Google Scholar
    • Export Citation
  • 37

    Leung GM, Rainer TH, Lau FL, Wong IO, Tong A, Wong TW, Kong JH, Hedley AJ, Lam TH; Hospital Authority SARS Collaborative Group, 2004. A clinical prediction rule for diagnosing severe acute respiratory syndrome in the emergency department. Ann Intern Med 141 :333–342.

    • Search Google Scholar
    • Export Citation
  • 38

    Watt G, Jongsakul K, Chouriyagune C, Paris R, 2003. Differentiating dengue virus infection from scrub typhus in Thai adults with fever. Am J Trop Med Hyg 68 :536–538.

    • Search Google Scholar
    • Export Citation
  • 39

    Sonthayanon P, Chierakul W, Wuthiekanun V, Blacksell SD, Pimda K, Suputtamongkol Y, Pukrittayakamee S, White NJ, Day NP, Peacock SJ, 2006. Rapid diagnosis of scrub typhus in rural Thailand using polymerase chain reaction. Am J Trop Med Hyg 75 :1099–1102.

    • Search Google Scholar
    • Export Citation
  • 40

    Singhsilarak T, Leowattana W, Looareesuwan S, Wongchotigul V, Jiang J, Richards AL, Watt G, 2005. Short report: detection of Orientia tsutsugamushi in clinical samples by quantitative real-time polymerase chain reaction. Am J Trop Med Hyg 72 :640–641.

    • Search Google Scholar
    • Export Citation
  • 41

    Watt G, Chouriyagune C, Ruangweerayud R, Watcharapichat P, Phulsuksombati D, Jongsakul K, Teja-Isavadharm P, Bhodhidatta D, Corcoran KD, Dasch GA, Strickman D, 1996. Scrub typhus infections poorly responsive to antibiotics in northern Thailand. Lancet 348 :86–89.

    • Search Google Scholar
    • Export Citation
  • 42

    Watt G, Kantipong P, Jongsakul K, Watcharapichat P, Phulsuksombati D, Strickman D, 2000. Doxycycline and rifampicin for mild scrub-typhus infections in northern Thailand: a randomised trial. Lancet 356 :1057–1061.

    • Search Google Scholar
    • Export Citation
Save