Abstract
Background
Scrub typhus, an emerging rickettsial disease caused by Orientia tsutsugamushi, is a clinically important endemic disease on Taiwan.
Methods
From January 1, 2007, to December 31, 2013, 156 patients diagnosed with scrub typhus were admitted to Taitung MacKay Memorial Hospital. Demographic data, clinical features, laboratory results, and outcomes of patients were retrospectively analyzed.
Results
Among 156 cases, 150 survived (96%) and six died (4%). There were 111 males (71%) and 45 females (29%) with a mean age of 47.8 years. The most common clinical features were fever (100%), general malaise (77%), chill (74%), headache (59%), and eschar (55%). No seasonal pattern was observed, with peaks in November and January. The average time to defervescence after appropriate antibiotics for the 150 surviving patients was 2.45 days. The significant risk factors were a delay of initial appropriate antibiotics use within 24 hrs, increased C-reactive protein (CRP), and liver cirrhosis.
Conclusion
Scrub typhus, a growing and emerging disease, is considered in the differential diagnosis of fever of unknown origin, and its diagnosis may be missed if it is not considered owing to its nonspecific clinical presentation. It is important to have a high index of suspicion and to increase awareness in endemic areas. Prompt diagnosis and early treatment with appropriate antibiotics are vital.
Keywords: scrub typhus, eschar, antibiotics
Introduction
“Scrub typhus” is a chigger-borne disease caused by Orientia tsutsugamushi (O. tsutsugamushi), an obligate intracellular bacterium in the Rickettsiaceae family. It is an acute febrile illness with the characteristic findings of high fever, maculopapular rash, lymphadenopathy, eschar, headache, myalgia, cough, sore throat, and gastrointestinal symptoms.[1,2] It is easy to be misdiagnosed if it is not considered, and is a unique feature of infection in endemic areas. The literature of the data for scrub typhus in Eastern Taiwan is lacking. Thus, we conducted this retrospective study to analyze the clinical data, laboratory data, complications, and outcomes of 156 patients diagnosed with scrub typhus during a consecutive 7-year period, with a review of the relevant literature.
Methods
Institution and Patients
We retrospectively reviewed all scrub typhus cases at Taitung MacKay Memorial Hospital, a 500-bed teaching hospital in Eastern Taiwan, between January 1, 2007 and December 31, 2013, from the records of the hospital infection control department. Dermographic data, clinical features, complications, “chest X-ray (CXR)” findings, laboratory findings, and outcomes were analyzed. A detailed clinical examination, including a careful search for eschar, was conducted. A diagnosis of scrub typhus was required with at least one of the following criteria: (1) a fourfold rise in the indirect immunofluorescent antibody (IFA) test for O. tsutsugamushi measured in acute and convalescent paired sera, (2) an Immunoglobulin M (IgM) antibody titre against O. tsutsugamushi > 1:80, and/or (3) a positive polymerase chain reaction (PCR) at the Center for Disease Control on Taiwan.[3,4] A total of 51 patients were diagnosed by PCR, and the remainings were diagnosed by serology.
Definitions
A systolic blood pressure below 90 mmHg or a reduction of more than 40 mmHg from the baseline in the absence of other causes of hypotension were defined as septic shock.[2] Acute renal failure was defined as a relative increase of 50% of the baseline serum creatinine concentration. For patients with an unknown baseline serum creatinine concentration,acute renal failure was diagnosed if the patient had a urine output of less than 0.5 ml/kg/hr for more than 6 hrs after treatment.[5] When the partial oxygen pressure of arterial blood (PaO2) was below 60 mmHg, with a normal or low partial carbon dioxide pressure of arterial blood (PaCO2) in room air, this was diagnosed as acute respiratory failure. CXR positive was defined according to the radiographic interpretations of radiologists. The radiographic findings were classified into pulmonary involvement (reticulonodular opacities, air space consolidation, peribronchial infiltration, and pleural effusion) and cardiovascular involvement (cardiomegaly with pulmonary congestion).[5] Aseptic meningitis and meningoencephalitis were diagnosed by a cerebrospinal fluid study.[6] Blood cultures were obtained at the time of hospitalization. Defervescence was defined as a drop in body temperature to less than 38°C that remains below this level for 24 hrs.[2] The time to defervescence was defined as the duration between the start of antibiotics and the resolution of fever. The duration of hospitalization and in-hospital mortality rate were also documented.
Statistical Analysis
The age and time to defervescence after antibiotics were presented by the mean and standard deviation (SD). Owing to the small sample sizes of the deceased group (n = 6), continuous data were presented by the median with an interquartile range (IQR) and were tested with the nonparametric Mann-Whitney test to determine the difference between groups. Categorical data were presented by numbers with percentages, and were tested with Fisher’s exact test for the difference between groups. Univariable and multivariable logistic regression models were performed to identify the impact factors for the risk of mortality, and were summarized by the odds ratio (OR) with a corresponding 95% confidence interval (CI). The factors that had significant associations with mortality in the univariable logistic regression models were included in the multivariable logistic regression model by the forward conditional selection method. Statistical assessments were evaluated using a two-sided significance level of 0.05 using SAS 9.4 for Windows® (SAS Institute Inc., Cary, NC, USA).
Results
Clinical Findings
A total of 156 patients with scrub typhus were enrolled and analyzed. Of these patients, 150 survived (96.2%) and six died (3.8%), with a mean age of 47.8 years (with SD of 15.6 years), including 111 males (71.2%) and 45 females (28.8%). No seasonal pattern was observed in our study, with peaks in November and January (Figure 1).
Figure 1. The distribution of 156 cases of scrub typhus by month of diagnosis.
Table 1 summarizes the clinical data at initial presentation. Common symptoms included fever, chills, general malaise, headache, cough, myalgia, arthralgia, eschar, and maculopapular rash. Septic shock was noted in 22 (14.1%) of the patients, acute renal failure in 12 (7.7%), acute respiratory failure in 8 (5.1%), and congestive heart failure (CHF) in 7 (4.4%). Central nervous system manifestations were not uncommon. Headache was the most common for 92 patients (59%), followed by confusion in 14 (9%) and seizure in 7 (4%) patients. Aseptic meningitis or meningoencephalitis was diagnosed in 12 (7.7%) patients. Gastrointestinal and hepatic involvement was common. A total of 18 patients without any recent history of peptic ulcer disease complained of epigastric pain, received a panendoscopy, and were observed to have a peptic ulcer with or without bleeding. Acute cholecystitis was noted in 15 (9.6%) patients. All of the six deceased patients were CXR positive, but only 45.3% of the surviving patients were CXR positive (p = 0.010). About half of our patients (49%) had a cough and had evidence of interstitial pneumonitis in their chest roentgenograms. The most frequently CXR abnormalities were diffuse, bilateral, and peribronchial infiltration. Pleural effusion was not uncommon, and 16 patients (21.6%) was presented with pleural effusion (Table 2).
Table 1. Clinical manifestations of the 156 scrub typhus patients.
|
Clinical manifestations |
Number (%) |
|
Symptoms |
|
|
Fever |
156 (100) |
|
General malaise |
120 (77) |
|
Chills |
116 (74) |
|
Headache |
92 (59) |
|
Cough |
76(49) |
|
Myalgia/arthralgia |
47 (30) |
|
Dyspnea |
36 (23) |
|
Abdomen pain |
36 (23) |
|
Nausea and vomit |
31(20) |
|
Dizziness |
17 (11) |
|
Confusion |
14 (9) |
|
Seizure |
7 (4) |
|
Signs |
|
|
Eschar |
86 (55) |
|
Maculopapular rash |
45 (29) |
|
Jaundice |
22(14) |
|
Hepatosplenomegaly |
22 (14) |
|
Lymphadenopathy |
16 (10) |
|
Neck stiffness |
5 (3) |
Table 2. Abnormal chest radiographic findings in 74 scrub typhus patientsa.
ARDS: acute respiratory distress syndrome.
aIndividual patients may have had more than one features.
|
Radiographic findings |
n (%) |
Distribution |
Zone |
||
|
bilateral |
unilateral |
upper |
lower |
||
|
Pulmonary involvement |
|
|
|
|
|
|
Reticulonodular opacities |
8/74 (10.8) |
8 |
0 |
0 |
8 |
|
Air space consolidation |
17/74 (22.9) |
10 |
7 |
3 |
14 |
|
Peribronchial infiltration |
35/74 (47.2) |
32 |
3 |
0 |
35 |
|
Pleural effusion |
16/74 (21.6) |
11 |
5 |
|
|
|
ARDS |
5/74 (6.7) |
5 |
0 |
|
|
|
Cardiovascular involvement |
|
|
|
|
|
|
Cardiomegaly with pulmonary edema |
7/74 (9.4) |
7 |
0 |
|
|
Location of Eschar
An eschar was present in 86 (55.1%) of the patients, with common sites including the groin and perineum (16.2%), lower extrimities (16.2%), axilla (13.9%), upper extremities (11.6%), back (10.4%), and abdomen (9.3%) (Table 3). In male patients, 62/111 (55.9%) of the patients presented with eschar, and for female patients, 24/45 (53.3%) of the patients presented with eschar. The probability of finding eschar during a physical examination seems not significantly different between males and females.
Table 3. Eschar distributions of scrub typhus cases.
|
Location of eschar |
Total number (%) |
Right |
Left |
Male |
Female |
|
Groin and perineum |
14 (16.2) |
5 |
9 |
13 |
1 |
|
Lower extrimities |
14 (16.2) |
7 |
7 |
8 |
6 |
|
Axilla |
12 (13.9) |
9 |
3 |
9 |
3 |
|
Upper extrimities |
10 (11.6) |
5 |
5 |
5 |
5 |
|
Back |
9 (10.4) |
4 |
5 |
6 |
3 |
|
Upper abdomen |
8 (9.3) |
5 |
3 |
7 |
1 |
|
Lower abdomen |
8 (9.3) |
5 |
3 |
7 |
1 |
|
Chest wall |
6 (6.9) |
2 |
4 |
5 |
1 |
|
Head and neck |
4( 4.6) |
3 |
1 |
2 |
2 |
|
Buttock |
1 (1.2) |
0 |
1 |
0 |
1 |
|
Total |
86 (100) |
45 |
41 |
62 |
24 |
Laboratory Findings
An initial blood count revealed leukocytosis (total white count > 12 × 103/μL) in 21 of the 156 patients (13.5%), leucopenia (total white count < 4 × 103/μL) in 9 of 156 patients (5.8%), and thrombocytopenia (platelet count < 150 × 103/μL) in 76 of 156 patients (48.7%). Hemoglobin of < 10 mg/dL was noted in 5 of 156 patients (3.2%). Prothrombin and activated partial thromboplastin time (> 12 s and > 36 s, respectively) were prolonged in 14 (8.9%) and 16 (10.2%) of the 156 patients, respectively. Acute renal failure was diagnosed in 12 of the 156 patients (7.7%). Although the serum potassium remained normal in most cases, serum sodium was slightly low and was < 130 mg/dL in 26 of 156 patients (16.6%). In 47 cases (30.1%), the serum albumin level was below 3 g/dL, and proteinuria was noted in 95 cases (60.9%). A total bilirubin of > 2 mg/dL was found in 25 of 156 patients (16%).
Microbiological Findings and Treatment
Routine blood cultures of all patients revealed no growth since O. tsutsugamushi can only be grown on mouse or chicken embryos or tissue cultures. All patients were treated with minocycline or doxycycline. The average time to defervescence after antibiotics for the 150 survived patients was 2.45 days (with SD of 1.16 days).
Clinical Outcomes and Factors Predictive of Death
Of the 156 patients, 6 (3.8%) died, and 150 (96.2%) survived. There were 106 males (70.7%) and 44 females (29.3%) with a median age of 46.5 years (IQR of 37–58 years) in the surviving group, and five males (83.3%) and one female (16.7%) with a median age of 54 years (IQR of 48–76 years) in the deceased group, respectively. Comparisons between the surviving and deceased scrub typhus patients are summarized inTable 4. The deceased group had a significantly longer duration of illness before admission (11.0 vs. 4.0 days, p < 0.001) and a prolonged hospitalization time (10.0 vs. 6.0 days, p < 0.001). Anemia, leukocytosis, thrombocytopenia, prolonged prothrombin time and activated partial thromboplastin time, jaundice, hypoalbuminemia, lower renal function, hyperkalemia, and higher C-reactive protein (CRP) were significantly more common in the deceased group. Liver cirrhosis patients had a significantly higher mortality compared to non-cirrhosis patients (p < 0.001). There were no malignant patients in our study. All six deceased patients were CXR positive, but only 45.3% of the surviving patients were CXR positive (p = 0.010). The deceased group had a significantly lower early initiation of appropriate antibiotics within 24 hrs compared to the surviving group (50 vs. 86.7, p = 0.042).
Table 4. Differences between the survival and died patients of scrub typhus.
ALB: albumin; ALT: alanine aminotransferase; APTT: activated partial thromboplastin time; AST: aspartate aminotransferase; BUN: blood urea nitrogen; Cr: creatinine; CRD: chronic renal disease; CRP: C-reactive protein; CVD: cardiovascular disease; CXR: chest X-ray interpretation; DM: diabetes mellitus; HB: hemoglobin; HTN: hypertension; K: potassium; LC: liver cirrhosis; Na: sodium; PLT: platelet; PT: prothrombin time; TB: total bilirubin; WBC: white blood cell.
aData are presented by median with inter-quartile range except for categorical data are presented by number with percentage.
* p < 0.05 indicates a statistically significant difference between the two groups.
** p was presented as < 0.001 when its value was 0.000 (given to three decimal places).
|
|
Scrub typhus |
p |
|
|
Survival (n = 150) |
Death (n = 6) |
||
|
Age (year) |
46.5 (37.0, 58.0) |
54.0 (48.0, 76.0) |
0.067 |
|
Gendera |
106 (70.7) |
5 (83.3) |
0.674 |
|
Male |
44 (29.3) |
1 (16.7) |
|
|
Female |
51 (34.0) |
3 (50.0) |
0.418 |
|
Hematuriaa |
91 (60.7) |
4 (66.7) |
1.000 |
|
Proteinuriaa |
68 (45.3) |
6 (100.0) |
0.010* |
|
CXR positivea |
85 (56.7) |
1 (16.7) |
0.090 |
|
Eschara |
44 (29.3) |
1 (16.7) |
0.674 |
|
Skin rasha |
46.5 (37.0, 58.0) |
54.0 (48.0, 76.0) |
0.067 |
|
Duration of illness before admission (day) |
4.0 (3.0, 7.0) |
11.0 (9.0, 14.0) |
< 0.001*,** |
|
Initiating of appropriate antibiotic within 24 hrs |
130 (86.7) |
3 (50.0) |
0.042* |
|
Comorbidity |
|
|
|
|
DM |
13 (8.7) |
0 |
1.000 |
|
HTN |
26 (17.3) |
3 (50.0) |
0.078 |
|
LC |
24 (16.0) |
5 (83.3) |
< 0.001*,** |
|
CVD |
19 (12.7) |
4 (66.7) |
0.005* |
|
CRD |
7 (4.7) |
2 (33.3) |
0.039* |
|
HB < 10 g/dL |
3 (2.0) |
2 (33.3) |
0.012* |
|
WBC > 10,000/µL |
32 (21.3) |
5 (83.3) |
0.003* |
|
PLT < 150,000/µL |
70 (46.7) |
6 (100.0) |
0.012* |
|
PT > 12 s |
8 (5.3) |
6 (100.0) |
< 0.001*,** |
|
APTT > 36 s |
10 (6.7) |
6 (100.0) |
< 0.001*,** |
|
AST > twice of normal U/L |
59 (39.3) |
4 (66.7) |
0.222 |
|
ALT > twice of normal U/L |
46 (30.7) |
3 (50.0) |
0.379 |
|
TB > 2 mg/dL |
19 (12.7) |
6 (100.0) |
< 0.001*,** |
|
ALB < 3 g/dL |
41 (27.3) |
6 (100.0) |
< 0.001*,** |
|
BUN > 20 mg/dL |
19 (12.7) |
6 (100.0) |
< 0.001*,** |
|
Cr > 2.0 mg/dL |
7 (4.7) |
5 (83.3) |
< 0.001*,** |
|
Na+ < 130 mEq/L |
26 (17.3) |
0 |
0.590 |
|
K+ > 5 mEq/L |
3 (2.0) |
3 (50.0) |
< 0.001*,** |
|
CRP > 10 mg/dL |
5 (3.3) |
4 (66.7) |
< 0.001*,** |
|
Hospitalization (day) |
6.0 (4.0, 8.0) |
10.0 (8.0, 13.0) |
0.014* |
In our study, the overall case fatality was 3.8%. According to the univariable logistic regression models, the risk of mortality increased with a longer duration before admission by an OR of 2.13 per day (p < 0.001). The risk of mortality increased with an increasing white cell count, creatinine, potassium, CRP, and comorbidities such as liver cirrhosis, cardiovascular disease, and chronic renal disease (with OR and 95% CI > 1), and decreased with the early initiation of appropriate antibiotics (minocycline or doxycycline) within 24 hrs (with OR and 95% CI < 1). In addition, given the associations of the three factors of early initiation of appropriate antibiotics within 24 hrs, liver cirrhosis and increased CRP remained statistically significant in the multivariable logistic regression model. After adjusting for each other, the risk of mortality decreased with the early initiation of appropriate antibiotics within 24 hrs (OR = 0.03 with 95% CI of 0–0.63, p = 0.023) and increased with increasing CRP (OR = 122.69 with 95% CI of 6.13–2,454.12, p = 0.002) and in liver cirrhotic patients (OR = 62.7 with 95% CI of 2.58–1,523.94, p = 0.011) (Table 5). A limitation of our study is the small sample size of the deceased group.
Table 5. Logistic regression models for the risk of mortality.
ALT: alanine aminotransferase; AST: aspartate aminotransferase; CI: confidence interval;Cr: creatinine; CRD: chronic renal disease; CRP: C-reactive protein; CVD: cardiovascular disease; CXR: chest X-ray interpretation; HB: hemoglobin; HTN: hypertension; K: potassium; LC: liver cirrhosis; OR: odds ratio; PLT: platelet; WBC: white blood cell. * p < 0.05 indicates a statistically significant association with mortality. ** p was presented as < 0.001 when its value was 0.000 (given to three decimal places).
|
Univariable |
Multivariable |
|||
|
Crude OR (95% CI) |
p |
Adjusted OR (95% CI) |
p |
|
|
Age (year) |
1.05 (1.00–1.11) |
0.068 |
||
|
Gender (female to male) |
0.48 (0.05–4.24) |
0.511 |
||
|
Hematuria |
1.94 (0.38–9.96) |
0.427 |
||
|
Proteinuria |
1.30 (0.23–7.30) |
0.768 |
||
|
CXR positive |
— |
|||
|
Eschar |
0.15 (0.02–1.34) |
0.090 |
||
|
Skin rash |
0.48 (0.05–4.24) |
0.511 |
||
|
Hospitalization (day) |
1.33 (1.06–1.67) |
0.015* |
||
|
Duration before admission (day) |
2.13 (1.36–3.32) |
< 0.001*,** |
||
|
Initiating of appropriate antibiotic within 24 hrs |
0.15 (0.03–0.82) |
0.028* |
0.03 (0.00–0.63) |
0.023* |
|
Comorbidity |
|
|
|
|
|
HTN |
4.77 (0.91–24.96) |
0.064 |
|
|
|
LC |
26.25 (2.93–234.77) |
0.003* |
62.70 (2.58–1,523.94) |
0.011* |
|
CVD |
13.79 (2.36–80.48) |
0.004* |
|
|
|
CRD |
10.21 (1.59–65.56) |
0.014* |
|
|
|
HB < 10 g/dL |
24.50 (3.17–189.63) |
0.002* |
||
|
WBC > 10,000/µL |
18.44 (2.08–163.48) |
0.009* |
||
|
PLT < 150,000/µL |
— |
|||
|
AST > twice of normal |
3.08 (0.55–17.38) |
0.202 |
||
|
ALT > twice of normal |
2.26 (0.44–11.63) |
0.329 |
||
|
Cr > 2.0 mg/dL |
102.11 (10.48–995.09) |
< 0.001*,** |
||
|
K > 5 mEq/L |
49.00 (6.86–350.16) |
< 0.001*,** |
||
|
CRP > 10 mg/dL |
58.00 (8.53–394.53) |
< 0.001*,** |
122.69 (6.13–2,454.12) |
0.002** |
Discussion
Scrub typhus is a zoonotic disease that is endemic to a part of the world known as the Tsutsugamushi Triangle. This extends from Japan to Taiwan, China, South Korea, Nepal, Northern Pakistan, Papua New Guinea, and the Australian states of Queensland and Northern New South Wales.[1,2,7] O. tsutsugamushi is a gram-negative, obligate intracellular bacterium that infects various cells, including endothelial cells and phagocytes, and causes acute vasculitis.[1] It is transmitted to humans by the bite of the larval thrombiculid mite, commonly known as a chigger, especially Leptotrombidium deliense on Taiwan. It is common in some parts of Taiwan, especially in Kinmen, Nantou, Penghu, Hualian, and Taitung counties, where our hospital is located. When humans are bitten by the infectious mites, rickettsia begins to grow at the location of the bite. Initially it forms a papule, which becomes a vesicle and then an ulcer, and which is finally covered by black eschar surrounded by red erythema. This forms after 9–12 days, and patients show clinical symptoms such as fever, maculopapular rash, headache, gastrointestinal symptoms, and lymphadenopathy. A total of 72% of patients with eschar were male, which may be a result of occupational exposure or outdoor activities in rural areas. Although the presence of an eschar is highly suggestive of scrub typhus, it was reported in variable proportions in various studies. Therefore, its presence confirms the disease, but its absence does not exclude the possibility. In a study in India, the absence of eschar was revealed as a poor prognostic factor because it was more difficult to make a diagnosis.[3,8,9] Five of six deceased patients had no eschar in our series, but no significant p value was noted. In a previous report, researchers used an eschar PCR for useful diagnostic tests on patients with scrub typhus. The nested PCR method for eschar might be both a rapid diagnostic test for tsutsugamushi disease in the early acute stage and a differential test for a crust is a scrub typhus eschar.[10] Scrub typhus is a notifiable disease in Taiwan, and the Center for Disease Control does not perform eschar PCR for the diagnosis of scrub typhus.
Fever is frequent in scrub typhus cases, with fever noted in 100% of the patients in our study. The systemic involvement of various organs results in complications such as meningoencephalitis, acute renal failure, myocariditis, interstitial pneumonia, adult respiratory distress syndrome (ARDS), and even multiple organ failure. Elevated aspartate aminotransferase, alanine aminotransferase, and bilirubin are common. Scrub typhus may be frequently overlooked as liver and biliary tract infections. Clinicians should be alert to the possibility that scrub typhus is intrinsically resistant to common antibiotics. Only effective appropriate antibiotic treatment can reduce the morbidity and mortality rates. Aseptic meningitis and meningoencephalitis are serious complications of scrub typhus. We found 12 cases (7.7%) of central nervous system involvement in our series. The findings of slight pleocytosis with mild protein elevation and normal sugar values in the cerebrospinal fluid in our cases are similar to those found in other studies.[6,11]
Radiological abnormalities in scrub typhus are not rare (74/156, 47.4%), and the findings vary. Reticulonodular opacities were found in eight patients (10.8%), airspace consolidation in 17 patients (22.9%), and peribronchial infiltrates in 35 patients (47.2%). In previous studies, pleural effusion was a common radiographic feature in 12–43% of patients. [12,13] Pleural effusion was found in 16 patients (21.6%) in our study, mostly combined with other parenchymal lesions. It is important to note that ARDS may develop in scrub typhus, and to be aware of the possible risk factors, because it is treatable if considered and diagnosed early. A total of 5 of 156 patients were found to have ARDS in our study. Park et al.[14] reported on pathological findings of open lung biopsies that showed interstitial edema with mononuclear cell infiltration, mainly lymphocytes without any evidence of vasculitis or perivasculitis. Thus, not only vasculitis owing to direct invasion of the rickettsial organism but also an immunological mechanism may participate in the pathogenesis of the pulmonary involvement of scrub typhus.[5,14-16] No autopsy or openlung biopsy was done in our series.
Cardiovascular complication has rarely been described in the past. Symptoms of myocardial involvement are usually subclinical. However, acute fulminant myocarditis leading to severe cardiogenic shock and fatalities was found in other studies. Cardiomegaly could be secondary to myocardial and pericardial involvement owing to the infection, and scrub typhus myocarditis may contribute to CHF. Acute onset of CHF was found in seven patients (9.4%) in our study, although no myocarditis was evaluated.[17,18]
In our study, scrub typhus had a predilection for males (male 111/156, 71.1%, female 45/156, 28.8%). This is similar to what has been reported in other series, and is assumed to result from the fact that men are the major workers and participants in field activities in Taiwan, as in many other countries. Thus, an increase in the risk of acquiring scrub typhus was noted in males. The high risk groups were the elderly, farmers, and males.
In general, hypoalbuminemia, thrombocytopenia, hepatic impairment, and acute renal insufficiency are associated with complications and mortality in patients with acute infectious diseases. In previous studies, a decreased albumin level, prolonged prothrombin time, and elevated creatinine levels were found to be predictors of mortality.[19,20] Although the mechanism of hypoalbuminemia in patients with scrub typhus is still not well known, it is considered to be associated with plasma protein leakage from the blood vessels owing to increased vascular permeability.[4,21-23] A delay of initial appropriate antibiotics use within 24 hrs, increased CRP, and liver cirrhosis were found to be our independent predictors of mortality.
In vitro, O. tsutsugamushi is susceptible to doxycycline and minocycline. In previous studies, a higher Acute Physiology and Chronic Health Evaluation (APACHE) II score was linked to high mortality for scrub typhus, and nearly all patients (93.6%) became afebrile within 72 hrs after the use of tetracycline antibiotics.[24,25] In our study, the average time to defervescence after antibiotics for the 150 surviving patients was 2.45 days (with SD of 1.16 days). The time to start tetracycline antibiotics after admission was the most important factor for scrub typhus patients.
Although scrub typhus is rare in pregnant females, when present, it can lead to serious repercussions for both the mother and fetus. The clinical features of pregnant females were the same as for non-pregnant females. The pregnancy outcome was closely related to the therapeutic outcome of each patient. Stillbirths and abortions were mainly observed in females whose scrub typhus was poorly controlled. Therefore, effective and safe therapeutic regimens are critical for avoiding adverse pregnancy outcomes. Kim et al. reported on eight pregnant women with scrub typhus who were treated successfully with a single 500-mg dose of azithromycin. All of the women delivered healthy babies at term without congenital or neonatal complications.[26] Since this is uncommon, no pregnant women were noted in our study.
In a study by Tsai and Yeh,[27] no significant correlation between scrub typhus incidence and surface temperature was observed in the endemic southeastern regions of Taiwan, including Taitung, whereas a similar significant correlation was found in low-incidence areas of the central-western and southwestern regions of Taiwan.[27,28] Our findings show that cases were found year-round and exhibited no climatic effects. However, the number of incidences during the cold season was higher than during the warm season. Peaks occurred in November and January in our study. This weather is suitable for outdoor activities, thus increasing the contact between infected mites and human hosts.
To improve self-recognition and encourage early visits to the hospital, it is important to provide intensive health education to local residents in endemic areas in order to increase awareness of clinical symptoms such as fever, eschar, headache, and skin rash. To reduce the risk of infection during epidemic periods, people should also be alerted to apply repellents to skin and clothes, to avoid contact with overgrown grasses and bushes, and to wear long pants, long sleeves, and long socks. Taiwan has a well-established public health system, so door-to-door health education by local public health nurses will be effective for prevention. We suggest that better vector control can also reduce the incidence of the disease.[29]
Although the rates of fatality have shown a considerable variation of about 14–17% between studies, the mortality rate was only 3.8% in our series. This might result from the high awareness of clinical physicians and thorough physical examinations of febrile patients in endemic areas. A limitation of our study was the small sample size of the deceased group. Further and larger studies will help to clarify more information regarding scrub typhus fever.
Scrub typhus, a growing and emerging disease, is considered a differential diagnosis of fever of unknown origin, and diagnosis may be missed if it is not considered owing to its nonspecific clinical presentations. Laboratory findings and clinical courses may also vary significantly, making diagnosis and appropriate treatment difficult. The possibility of scrub typhus, which is intrinsically resistant to common empirical antibiotics, should be kept in mind with regard to fever patients with abnormal liver function and jaundice. It may be frequently overlooked as a biliary tract infection. Travel history to endemic areas of Taiwan is also important in these patients. Our retrospective study highlights that a delayed diagnosis and treatment may be fatal. Therefore, it is important to have a high index of suspicion and to increase awareness in endemic areas at initial presentation. Prompt diagnosis and early treatment with appropriate antibiotics are crucial for patient survival.
Conflicts of Interest Statement
None to declare.
Declaration of Interest
There are no commercial relationship or potential conflict of interest, no sources of funding and the study was not supported by any institutions.
References
- 1.Varghese George M., Janardhanan Jeshina, Trowbridge Paul, Peter John V., Prakash John A.J., Sathyendra Sowmya, Thomas Kurien, David Thambu S., Kavitha M.L., Abraham Ooriapadickal C., Mathai Dilip. Scrub typhus in South India: clinical and laboratory manifestations, genetic variability, and outcome. International Journal of Infectious Diseases. 2013 Nov;17(11) doi: 10.1016/j.ijid.2013.05.017. [DOI] [PubMed] [Google Scholar]
- 2.Lee H C, Ko W C, Lee H L, Chen H Y. Clinical manifestations and complications of rickettsiosis in Southern Taiwan. J Formos Med Assoc. 2002;101:385–392. [PubMed] [Google Scholar]
- 3.Kim D M, Won K J, Park C Y, et al Distribution of eschars on the body of scrub typhus patients: a prospective study. Am J Trop Med Hyg. 2007;76:806–809. [PubMed] [Google Scholar]
- 4.Liang Xu-jing, Huang Si-min, Li Jian-ping, Zhu Xian-nv, Lu Yong-ping, Hocher Berthold, Chen You-peng. Hepatic Impairment Induced by Scrub Typhus is Associated with New Onset of Renal Dysfunction. Clinical Laboratory. 2014;60(01/2014) doi: 10.7754/clin.lab.2013.121203. [DOI] [PubMed] [Google Scholar]
- 5.Hsu Yung-Hsiang, Chen Hsing I. Pulmonary pathology in patients associated with scrub typhus. Pathology. 2008 Apr;40(3) doi: 10.1080/00313020801911488. [DOI] [PubMed] [Google Scholar]
- 6.Lee W S, Lee W C, Chen T L, Fung C P. Scrub typhus complicating interstitial pneumonitis and meningoencephalitis: a case report Thorac Med. 2004;19:50–54. [Google Scholar]
- 7.Koh Gavin C. K. W., Paris Daniel H., Blacksell Stuart D., Maude Richard J., Newton Paul N. Diagnosis of Scrub Typhus. The American Journal of Tropical Medicine and Hygiene. 2010 Mar 01;82(3) doi: 10.4269/ajtmh.2010.09-0233. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bhattacharyya Prithwis, Rapsang AmyG. Scrub typhus. Indian Journal of Anaesthesia. 2013;57(2) doi: 10.4103/0019-5049.111835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zhang Meng, Wang Xian-Jun, Ding Lei, Zhao Zhong-Tang, Li Zhong, Ding Shu-Jun. Scrub Typhus: Surveillance, Clinical Profile and Diagnostic Issues in Shandong, China. The American Journal of Tropical Medicine and Hygiene. 2012 Dec 05;87(6) doi: 10.4269/ajtmh.2012.12-0306. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lee S.-H., Kim D.-M., Cho Y. S., Yoon S. H., Shim S. K. Usefulness of Eschar PCR for Diagnosis of Scrub Typhus. Journal of Clinical Microbiology. 2006 Mar 01;44(3) doi: 10.1128/jcm.44.3.1169-1171.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Viswanathan Stalin, Muthu Vivekanandan, Iqbal Nayyar, Remalayam Bhavith, George Tarun. Paul Friedemann., editor. Scrub Typhus Meningitis in South India — A Retrospective Study. PLoS ONE. 2013 Jun 14;8(6) doi: 10.1371/journal.pone.0066595. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Acute Respiratory Distress Syndrome Following Scrub Typhus 胸腔醫學. 2009 Aug 01;24(4) doi: 10.29806/TM.200908.0005. [DOI] [Google Scholar]
- 13.Charoensak A, Chawalparit O, Suttinont C, et al Scrub typhus: chest radiographic and clinical findings in 130 Thai patients. J Med Assoc Thai. 2006;89:600–607. [PubMed] [Google Scholar]
- 14.Park Jae Seuk, Jee Young Koo, Lee Kye Young, Kim Keun Yeol, Myong Na Hye, Seo Pil Weon. Acute respiratory distress syndrome associated with scrub typhus: diffuse alveolar damage without pulmonary vasculitis. Journal of Korean Medical Science. 2000;15(3) doi: 10.3346/jkms.2000.15.3.343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Pandey D, Sharma B, Chauhan V, Mokta J, Verma B S, Thakur S. ARDS complicating scrub typhus in sub-Himalayan region. J Assoc Physicians India. 2006;54:812–813. [PubMed] [Google Scholar]
- 16.Jeong Yeon Joo, Kim Suk, Wook Yeh Dae, Lee Jun Woo, Kim Kun-Il, Lee Sun Hee. Scrub Typhus: Clinical, Pathologic, and Imaging Findings. RadioGraphics. 2007 Jan;27(1) doi: 10.1148/rg.271065074. [DOI] [PubMed] [Google Scholar]
- 17.Sittiwangkul Rekwan, Pongprot Yupada, Silviliarat Suchaya, Oberdorfer Peninnah, Jittamala Podjanee, Sirisanthana Virat. Acute fulminant myocarditis in scrub typhus. Annals of Tropical Paediatrics. 2008 Jun;28(2) doi: 10.1179/146532808x302189. [DOI] [PubMed] [Google Scholar]
- 18.Jung Hyun Chang, Mi Soon Ju, Jee Eu Pericarditis due to Tsutsugamushi Disease. Scandinavian Journal of Infectious Diseases. 2000 Jan;32(1) doi: 10.1080/00365540050164344. [DOI] [PubMed] [Google Scholar]
- 19.VARGHESE G, ABRAHAM O, MATHAI D, THOMAS K, AARON R, KAVITHA M, MATHAI E. Scrub typhus among hospitalised patients with febrile illness in South India: magnitude and clinical predictors. Journal of Infection. 2006 Jan;52(1) doi: 10.1016/j.jinf.2005.02.001. [DOI] [PubMed] [Google Scholar]
- 20.Wang C C, Liu S F, Liu J W, Chung Y H, Su M C, Lin M C. Acute respiratory distress syndrome in scrub typhus. J Trop Med Hyg. 2007;76:1148–1152. [PubMed] [Google Scholar]
- 21.Hu M L, Liu J W, Wu K L, et al Short report: abnormal liver function in scrub typhus. Am J Trop Med Hyg. 2005;73:667–668. [PubMed] [Google Scholar]
- 22.Narvencar K P S, Rodrigues S, Nevrekar R P, et al Scrub typhus in patients reporting with acute febrile illness at a tertiary health care institution in Goa. Indian J Med Res. 2012;136:1020–1024. [PMC free article] [PubMed] [Google Scholar]
- 23.Lee C S, Hwang J H, Lee H B, Kwon K S. Risk factors leading to fatal outcome in scrub typhus patients. Am J Trop Med Hyg. 2009;81:484–488. [PubMed] [Google Scholar]
- 24.Lee Chang-Seop, Min In-Suk, Hwang Jeong-Hwan, Kwon Keun-Sang, Lee Heung-Bum. Clinical significance of hypoalbuminemia in outcome of patients with scrub typhus. BMC Infectious Diseases. 2010 Jul 21;10(1) doi: 10.1186/1471-2334-10-216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Tsai Chen-Chi, Lay Chorng-Jang, Ho Yu-Huai, Wang Lih-Shinn, Chen Li-Kuang. Intravenous minocycline versus oral doxycycline for the treatment of noncomplicated scrub typhus. Journal of Microbiology, Immunology and Infection. 2011 Feb;44(1) doi: 10.1016/j.jmii.2011.01.007. [DOI] [PubMed] [Google Scholar]
- 26.Kim Y S, Lee H J, Chang M, Son S K, Rhee Y E, Shim S K. Scrub typhus during pregnancy and its treatment: a case series and review of the literature. Am J Trop Med Hyg. 2006;75:955–959. [PubMed] [Google Scholar]
- 27.Tsai Pui-Jen, Yeh Hsi-Chyi. Scrub typhus islands in the Taiwan area and the association between scrub typhus disease and forest land use and farmer population density: geographically weighted regression. BMC Infectious Diseases. 2013 Apr 29;13(1) doi: 10.1186/1471-2334-13-191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kuo Chi-Chien, Huang Jing-Lun, Ko Chia-Ying, Lee Pei-Fen, Wang Hsi-Chieh. Spatial analysis of scrub typhus infection and its association with environmental and socioeconomic factors in Taiwan. Acta Tropica. 2011 Oct;120(1-2) doi: 10.1016/j.actatropica.2011.05.018. [DOI] [PubMed] [Google Scholar]
- 29.Sharma Puran K., Ramakrishnan R., Hutin Y.J.F., Barui A.K., Manickam P., Kakkar M., Mittal V., Gupte M.D. Scrub typhus in Darjeeling, India: opportunities for simple, practical prevention measures. Transactions of the Royal Society of Tropical Medicine and Hygiene. 2009 Nov;103(11) doi: 10.1016/j.trstmh.2009.02.006. [DOI] [PubMed] [Google Scholar]