Our study addresses the sepsis research gap in lower middle-income countries, notably India. Here, we investigate community-acquired sepsis comprehensively and explore the impact of tropical microbiology on aetiology and outcomes. MARS-India was a prospective observational study from Dec-2018 to Sep-2022 in a tertiary-care hospital in South India. Adult patients within 24hrs of ICU admission meeting the Sepsis 3.0 definition were enrolled, with 6-months follow-up (http://clinicaltrials.gov number NCT03727243). Over 4000 patients were screened on ICU admission, with 1000 unique patients meeting the inclusion criteria. Median age was 55 years (IQR: 44-65) with a male preponderance (66%). Almost half the cohort resided in villages (46.5%) and 74.6% worked in the primary sector. Mortality in-hospital was 24.1%. Overall, ∼54% had confirmed microbiological diagnosis. Over 18% had a viral cause of sepsis. Surprisingly, we identified leptospirosis (10.6%), scrub typhus (4.1%), dengue (3.7%) and Kyasanur forest disease (1.6%) as notables causes of sepsis. All these infections showed seasonal variation around the monsoon. In community-acquired infections we observed substantial resistance to 3rd generation cephalosporins and carbapenems. In India, sepsis disproportionally affects a younger and lower socio-economic demographic, yielding high mortality. Tropical and viral sepsis carry a significant burden. Analyzing local data, we pinpoint priorities for public health and resources, offering valuable insights for global sepsis research.
21 March 2026 in The Lancet Regional Health - Southeast Asia
Characterising the host response in sepsis is essential to understand its biological heterogeneity and to inform more precise diagnostic and therapeutic strategies. Existing evidence on sepsis comes predominantly from studies conducted in high-income countries (HICs), despite the highest burden in low- and middle-income countries (LMICs). We aimed to address this gap by identifying shared and pathogen-specific host response patterns among different infectious causes of sepsis in a tertiary care centre in India. Patients fulfilling sepsis-3 criteria were enrolled within 24 h of intensive care unit admission in a tertiary care centre in Manipal, Karnataka, India. We measured 27 plasma biomarkers reflecting key pathophysiological domains (endothelial activation and coagulation, organ damage and inflammation, cytokine response, and chemokine release) to delineate host response profiles across sepsis aetiologies. We included 956 sepsis patients, and a causative pathogen was identified in 54·1% (338 bacterial, 146 viral, 33 polymicrobial). The causative pathogen explained a significant proportion of biomarker variation (34·5% of explained variance; 9·4% of all variances). While bacterial sepsis was associated with stronger host response changes across all domains when compared to viral sepsis, notable variation was found within these microbial categories. Orientia tsutsugamushi (causing scrub typhus) triggered the most profound host response alterations among bacterial causes, whereas viral haemorrhagic fever (dengue, Kyasanur forest disease) causing pathogens induced higher biomarker levels across all pathophysiological domains compared to SARS-CoV-2 or influenza within viral sepsis. These findings highlight the biological heterogeneity of sepsis and the complexity of host–pathogen interactions in a setting with a diverse range of causative organisms. 10.13039/501100000780 European Union , 10.13039/100019573 Amsterdam UMC , and 10.13039/100019305 Manipal Academy of Higher Education .