Severe leptospirosis in the intensive care unit: a single-centre prospective cohort study from Sri Lanka

Description:This dataset contains de-identified clinical, laboratory, and outcome data from 116 patients with microbiologically confirmed severe leptospirosis admitted to the intensive care unit (ICU) at the National Hospital of Sri Lanka between 2019 and 2022. The dataset was used in a published cohort study which characterized phenotypic clusters of severe leptospirosis using unsupervised machine learning methods. It includes variables related to organ dysfunction, treatment modalities, and outcomes such as death and ICU length of stay. Files Included:- `Lepto_ICU_cleaned_for_Zenodo.csv`: Cleaned dataset (one row per patient)- `Lepto_ICU_Data_Dictionary.csv`: Data dictionary with variable names, types, and example values Key Variables:- Demographics: Age, Sex, Weight- Vital signs: SBP, DBP, MAP, HR (Admission and ICU values)- Laboratory: WBC, Hemoglobin, Platelets, CRP, Creatinine, AST, ALT, INR, APTT, Electrolytes, pH, Lactate- Complications: Pulmonary haemorrhage, Myocarditis, AKI, Coagulopathy, Sepsis, Shock- Interventions: MV, CRRT, SLED, PRC, FFP, PLEX, Steroids- Outcomes: ICU stay duration, Death Ethical Considerations:All data are fully de-identified. Ethics approval was obtained from the Ethics Review Committee of the National Hospital of Sri Lanka. The dataset is shared for academic, non-commercial reuse under the terms of the CC-BY 4.0 license. Citation:Please cite the dataset as:Ruwanpathirana P, et al. (2025). Severe Leptospirosis ICU Cohort – Sri Lanka [Data set]. Zenodo. https://doi.org/10.5281/zenodo.15709911 Contact:For questions or collaboration requests, please contact:Dr. Pramith Ruwanpathirana ([e-mail])