Introduction

Scrub typhus is a leading cause of acute febrile illness in the Asia-Pacific region, with a pooled seroprevalence of ~25%, and up to ~60% of individuals showing evidence of prior infection (IgG positivity) [1]. Scrub typhus is caused by Orientia tsutsugamushi, an obligate intracellular bacterium transmitted through the bites of infected chigger mites [2]. While scrub typhus is reported across many countries, the highest burden occurs within the “tsutsugamushi triangle” [3]. Global mortality rates range from 1.4% among treated patients to 6% among untreated individuals [3].

South Korea has been reporting a steady increase in the cases of scrub typhus after it has been designated as a notifiable disease by the Korea Diseases Control and Prevention Agency. Between 2001 and 2013, there were 70,914 cases reported [4]. The overall annual incidence rate of scrub typhus was 12.2 cases per 100,000 residents for the period 2008–2012 [5]. The trend of cases reported has continued to increase with risk factors such as outdoor activities [6], deforestation [7] and climatic factors such as temperature and precipitation [8].

Hantavirus infection is another significant cause of acute febrile illness, with a substantial caseload reported from the Asia-Pacific region [9]. Hantaviruses are zoonotic viruses, with rodents serving as their primary natural reservoirs. These viruses belong to the genus Orthohantavirus, family Hantaviridae, and order Bunyavirales. Common hantavirus variants reported in Asia include Hantaan virus (HTNV), Puumala virus (PUUV), Seoul virus (SEOV), Muju virus (MUJV), and Soochong virus (SOOV) [9]. Hantaviruses cause two major clinical syndromes: hemorrhagic fever with renal syndrome (HFRS) and hantavirus cardiopulmonary syndrome (HCPS). The global seroprevalence of hantavirus infections is 2.95%, with a higher prevalence of 6.84% reported in Asia [10]. Although mortality rates can reach 15% in some studies [10,11], global data on variant- or syndrome-specific mortality rates remain limited.

Hantaviruses are endemic to several Asian countries, such as China, South Korea and Japan [12]. Hantavirus was first identified in a soldier stationed in the center front, known as the Iron Triangle [13]. Between 2001 and 2017, there were 7048 cases of HFRS and 31 deaths reported in South Korea with an annual incidence of 0.83 per 100,000 residents and a case fatality rate of 1.26% [14]. The majority of HFRS cases occur during seasonal peak noted between October and December [13].

Scrub typhus and hantavirus infections share overlapping clinical presentations, including acute febrile illness and thrombocytopenia, and they are endemic in similar geographic regions. HFRS progresses through five clinical stages: febrile, hypotensive, oliguric, diuretic, and convalescent phases [9]. Similarly, severe scrub typhus can present with organ dysfunction, including shock, acute kidney injury, respiratory failure, and multiorgan dysfunction [15]. This study compares the clinical, laboratory, and epidemiological characteristics of scrub typhus and HFRS, using molecular and serological diagnostic confirmation, in patients treated at a tertiary-care teaching hospital in South Korea.

Methods

Results

A total of 207 patient records were reviewed. Six cases of co-infection with scrub typhus and HTNV were excluded. Of the remaining cases, 156 (77.61%) patients were diagnosed with scrub typhus, and 45 (22.39%) were diagnosed with HFRS.

Among patients with HFRS, 33 (76.74%) tested positive via PCR, and 31 (68.89%) demonstrated a four-fold increase in IgG titers.

Among patients with scrub typhus, the Orientia tsutsugamushi strain distribution was as follows: 136 (87.18%) were Boryong strain, 4 (2.56%) were Taguchi strain, 3 (1.93%) were Karp strain, and 1 (0.64%) was Kanda strain. Nine specimens could not be sequenced, and sequencing was not performed for three specimens.

Sociodemographic, clinical and laboratory parameters

The mean age of patients with scrub typhus was 69.52 ± 11.25 years, with a majority aged ≥65 years (103, 66.03%). In contrast, the mean age of patients with HFRS was 56.84 ± 17.24 years. The detailed sociodemographic characteristics are presented in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Sociodemographic, clinical and laboratory characteristics of patients with scrub typhus infection and hemorrhagic fever with renal syndrome (HFRS) treated at the Chosun University Hospital, South Korea, 2010–2023.

https://doi.org/10.1371/journal.pntd.0013709.t001

The mean duration of hospitalization was significantly shorter for patients with scrub typhus (7.89 ± 7.56 days) compared to patients with HFRS (14.09 ± 7.67 days; p < 0.001). The median hospital stays were 6 days (IQR 4, 8) for patients with scrub typhus and 12 days (IQR 8, 21) for patients with HFRS. A higher proportion of patients with HFRS (15 patients, 33.33%) required intensive care admission compared to those with scrub typhus (14 patients, 8.97%), p < 0.001. Eleven patients with scrub typhus requiring ICU admission had Boryong strain, one had Taguchi strain and two did not have sequencing information.

There were 5 deaths (3.21%) among patients with scrub typhus and 1 (2.22%) death among those with HFRS, p = 0.733. Three of the deaths among scrub typhus had Boryong strain, the other two did not have sequencing information. The clinical characteristics, hospitalization duration, and outcomes are detailed in Table 1.

Patients with scrub typhus had a lower median leukocyte count (8105/µL, IQR 5370, 11035) compared to patients with HFRS (14840/µL, IQR 7940, 21200; p < 0.001). Scrub typhus cases had a higher median platelet count (124000/µL, IQR 99000, 157000) compared to HFRS cases (55,000/µL, IQR 31000, 124000; p < 0.001). Patients with scrub typhus had a median AST 90.10 U/L (IQR 55.85, 139.35 U/L) while those with HFRS had a median AST 83.40 U/L (IQR 62.90, 183 U/L), p = 0.754. Patients with scrub typhus had a median ALT 63.50 U/L (IQR 38.10, 108.75 U/L) while those with HFRS had a median ALT 42.80 U/L (IQR 30.50, 83.30 U/L), p = 0.030. Patients with scrub typhus had a median creatinine 0.95 mg/dL (IQR 0.76, 1.34 mg/dL) while those with HFRS had a median creatinine 1.93 mg/dL (IQR 1.12, 4.80 mg/dL), p < 0.001. The detailed differences in laboratory parameters between patients with scrub typhus and HFRS are presented in Fig 1 and Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Distribution of (a) leukocyte and (b) platelet counts, and (c) neutrophil and (d) lymphocyte percentages in patients with hemorrhagic fever with renal syndrome (HFRS) and scrub typhus infection treated at the Chosun University Hospital, South Korea, 2010 – 2023.

https://doi.org/10.1371/journal.pntd.0013709.g001

Risk factors associated with scrub typhus

The sociodemographic, clinical and laboratory parameters associated with scrub typhus are presented in Table 2. Adjusted analysis identified the following significant associations with scrub typhus: rash (adjusted OR: 1.23, 95% CI: 1.10–1.37, p < 0.001), eschar (adjusted OR: 1.16, 95% CI: 1.06–1.26, p = 0.001), and lymphopenia (<1500/µL) (adjusted OR: 1.42, 95% CI: 1.16 – 1.74, p = 0.001).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Association of sociodemographic, clinical and laboratory parameters with scrub typhus infection compared to hemorrhagic fever with renal syndrome (HFRS) among patients treated at the Chosun University Hospital, South Korea, 2010–2023.

https://doi.org/10.1371/journal.pntd.0013709.t002

Discussion

Scrub typhus and HFRS present as acute febrile illnesses with overlapping but non-specific symptoms sometimes leading to misdiagnosis if not supported by reliable laboratory evidence [18]. In this study, the highest proportion of both scrub typhus and HFRS was observed in individuals aged ≥65 years. In a review of scrub typhus cases from 2001 to 2013 in South Korea, the highest incidence was reported in the 60–69-year age group [4]. This contrasts with a median age of 33 years reported in Vietnam [19] and 48 years in South India [20]. For HFRS, an earlier study in South Korea (2002–2012) found the highest proportion of cases in individuals aged >40 years [21]. In China, the at-risk age group for HTNV was 40–59 years [22]. In the United States, HCPS cases confirmed through serology or PCR, PUUV, SEOV, Prospect Hill virus, and SNV showed a median age of 34.9 years [23]. Similarly, HCPS caused by Andes virus in Argentina (1995–2008) predominantly affected individuals aged 21–30 years [24], while cases in Chile were reported in the 15–44-year age group [25].

In South Korea, individuals aged ≥65 years represent the main workforce engaged in outdoor and agricultural activities, particularly during the harvest season between October and December when exposure to chigger mites (Leptotrombidium spp.) and infected rodents is highest [26]. This demographic pattern reflects socio-economic and demographic shifts, as younger populations have increasingly migrated to urban areas, leaving older adults to perform farming and fieldwork in rural communities [6,27,28]. Consequently, older adults experience greater exposure to vector habitats, leading to a higher incidence of both Orientia tsutsugamushi (scrub typhus) and HTNV/SEOV infections. Additionally, the rodent population density and virus shedding rates rise after the breeding season in late summer and early autumn, further elevating the risk of human infection [11]. These findings are consistent with previous national epidemiological data showing that most scrub typhus and HFRS cases in Korea occur among elderly farmers during the autumn season [27,28].

A higher proportion of scrub typhus cases occurred in females. A global review of epidemiological studies identified females as an at-risk population for scrub typhus [3], with slightly higher disability-adjusted life years compared to males [29]. Conversely, HFRS was more prevalent among males, consistent with a reported male-to-female ratio of 2.6:1 in adults, while the ratio was 1:1 among children [9]. The observed sex and age distributions in scrub typhus and HFRS may be explained by differences in human activities, occupational exposure, and season-related risk-factors.

Fever is a hallmark symptom of scrub typhus [5,30], reported in nine out of ten patients in this study. While a febrile phase lasting 4–6 days is typically described in HFRS [31], only half of the patients in our study reported fever, in contrast to the 86.9-97.1% reported in studies from Korea and China [22,32] and 100% in critical ill cases [33]. In cases with high-risk exposure to rodent carriers, Hantavirus infection should be considered a differential diagnosis even in the absence of fever. Severe cases of HTNV or Dobrava virus (DOBV) infection typically exhibit five clinical phases, but these may not be apparent in less severe forms caused by PUUV or SEOV [11,34]. Prospective studies with detailed clinical descriptions are recommended to characterize variant-specific symptoms of Hantavirus infections.

In scrub typhus, diffuse, macular or maculopapular rashes appear 3–10 days after fever onset [30], while eschars may develop up to one week before fever [35]. In our study, patients with scrub typhus were more likely to have skin rashes and eschars compared to those with HFRS. Eschars may be a diagnostic clue but may not always be present. The proportion of eschars observed in this study was consistent with reports of 78.9-80.4% in Korea [5,36]. However, eschars have been reported globally in only 30.3% of cases, with lower rates in South Asia [37]. Despite variability, head-to-toe clinical examinations, including areas under compression or restrictive clothing, remain essential [38].

Higher proportions of hemorrhagic symptoms and severe thrombocytopenia (platelets <50,000/µL) were observed in HFRS cases. Bleeding manifestations included petechiae, epistaxis, menorrhagia, and gastrointestinal bleeding. While fatal bleeding is rare, it is more common in HTNV and DOBV infections [9]. The primary mechanisms of hemorrhage in HFRS are increased vascular permeability and thrombocytopenia, which also contribute to renal impairment [11]. Renal injury results from tubular damage caused by cell infiltration and increased glomerular permeability, leading to massive proteinuria [11].

In our study, leukocytosis was more frequently observed in HFRS cases compared to cases with scrub typhus. Hantavirus infections are associated with elevated leukocyte populations [39,40] and a marked increase in activated T cells, which contribute to cytokine storms [41]. This immune cascade, involving natural killer and T cell activation, correlates with viremia and may offer targets for therapeutic interventions [42].

Conversely, in our study, scrub typhus was associated with lymphocytopenia (<1500/µL) during the acute phase, with counts returning to baseline during convalescence [43]. This phenomenon may result from CD4+ and CD8+ T cell apoptosis during the acute phase, followed by CD8+ T cell proliferation in the convalescent phase. In hantavirus infections, increased percentages of CD8+ T cells with activation markers (e.g., CD71 and CD25) correlate positively with AST and ALT levels and a negatively with platelet counts [44]. Further research is warranted to elucidate these immune mechanisms.

Scrub typhus can cause multiorgan dysfunction, including respiratory distress, acute kidney injury, and meningoencephalitis [30]. Bleeding manifestations occur in up to 6.2% of scrub typhus cases [37]. Disease severity is influenced by host factors, pathogen factors (e.g., Orientia tsutsugamushi strain) [45], and bacterial load in specific organs [46]. Strains such as Karp, Gilliam, and Woods have been associated with higher levels of endothelial cell activation markers [46], which may explain variations in clinical severity. Given the overlapping clinical features of scrub typhus and HFRS, further studies are required to understand their full clinical spectrum.

In our study, patients with HFRS had relatively longer hospital stays and higher rates of ICU admission compared to those with scrub typhus. The duration of hospitalization for scrub typhus was consistent with reports from other countries [20]. In a review of 114 patients with HFRS, the average duration of hospital stay was 10.98 ± 4.81 days among non-pregnant female patients and 11.15 ± 4.41 days among pregnant patients [47]. In another review of 35 patients with HFRS, the average duration of hospital stage was 11.7 ± 5.3 days [32]. Patients at our hospital had a slightly longer stay in hospital that may reflect disease severity, delays in diagnosis, or low prioritization of HTNV testing.

This study was based on a retrospective review of data from a single teaching hospital. Comparisons of clinical and laboratory parameters are conducted only for PCR-confirmed cases. While a few cases might have been missed in instances where tests were not offered, the number might be negligible given the relatively easy access to molecular testing for diagnosis. Given that scrub typhus and HFRS have overlapping clinical findings and that only a small number of cases with HFRS were captured in this study, further prospective multicenter studies are necessary to validate the differences in clinical and laboratory parameters presented here.

Conclusions

In our study, patients with scrub typhus and HFRS presented with overlapping clinical features and had similar mortality rates. However, patients with skin rash, eschar and lymphopenia were more likely to have scrub typhus. The duration of hospital stay was longer in patients with HFRS. Further studies are recommended to describe the clinical features and disease course in patients with various subtypes of Hantavirus infections.

Supporting information