Neurological manifestations of scrub typhus infection: A systematic review and meta-analysis of clinical features and case fatality

Abstract

Background: Scrub typhus has become a leading cause of central nervous system (CNS) infection in endemic regions. As a treatable condition, prompt recognition is vital. However, few studies have focused on describing the symptomology and outcomes of neurological scrub typhus infection. We conducted a systematic review and meta-analysis to report the clinical features and case fatality ratio (CFR) in patients with CNS scrub typhus infection.

Methods: A search and analysis plan was published in PROSPERO [ID 328732]. A systematic search of PubMed and Scopus was performed and studies describing patients with CNS manifestations of proven scrub typhus infection were included. The outcomes studied were weighted pooled prevalence (WPP) of clinical features during illness and weighted CFR.

Results: Nineteen studies with 1,221 (656 adults and 565 paediatric) patients were included. The most common clinical features in CNS scrub typhus were those consistent with non-specific acute encephalitis syndromes (AES), such as fever (WPP 100.0% [99.5%-100.0%, I2 = 47.8%]), altered sensorium (67.4% [54.9-78.8%, I2 = 93.3%]), headache (65.0% [51.5-77.6%, I2 = 95.1%]) and neck stiffness 56.6% (29.4-80.4%, I2 = 96.3%). Classical features of scrub typhus were infrequently identified; an eschar was found in only 20.8% (9.8%-34.3%, I2 = 95.4%) and lymphadenopathy in 24.1% (95% CI 11.8% - 38.9%, I2 = 87.8%). The pooled CFR (95% CI) was 3.6% (1.5%- 6.4%, I2 = 67.3%). Paediatric cohorts had a CFR of 6.1% (1.9-12.1%, I2 = 77%) whilst adult cohorts reported 2.6% (0.7-5.3%, I2 = 43%).

Conclusion: Our meta-analyses illustrate that 3.6% of patients with CNS manifestations of scrub typhus die. Clinicians should have a high index of suspicion for scrub typhus in patients presenting with AES in endemic regions and consider starting empiric treatment whilst awaiting results of investigations, even in the absence of classical signs such as an eschar or lymphadenopathy.

Fig 1. Study selection.

Fig 2. Quality assessment for the risk of bias in the 19 included studies for meta-analyses.

Fig 3. Forest plot showing case fatality ratio for adult, paediatric and combined cohorts; events are defined as number of in hospital fatalities [,–,–,–26,28,29,37].

Fig 4

A: Map illustrating tsustugamushi triangle and the included studies in our meta-analyses, circles are proportionate to the size of the cohorts. The map was created using public domain image (https://commons.wikimedia.org/wiki/File:BlankMap-World.svg). B: Graph illustrating number of published cases of central nervous system (CNS) scrub typhus described in literature.