Multicentre open-label randomised controlled trial to evaluate the efficacy and safety of moderate-dose and high-dose intravenous methylprednisolone combined with plasmapheresis versus plasmapheresis alone in leptospirosis pulmonary haemorrhage syndrome: a protocol

Abstract

Introduction

Leptospirosis is a zoonotic infection caused by the bacterium Leptospira, which typically enters humans through contact with water or soil contaminated with infected urine. One of the most important causes of death due to leptospirosis is leptospirosis pulmonary haemorrhage syndrome (LPHS). LPHS occurs due to an inappropriate immune response, resulting in excessive release of cytokines/chemokines or direct damage to alveoli by bacteria. Mortality remains high in LPHS, and management strategies of this deadly complication are under-investigated. Survivors have minimal residual lung injury, necessitating evidence-based treatment to improve mortality in LPHS. Steroids and plasmapheresis have been tried in many case series but lack robust evidence. Plasmapheresis has become a standard practice in Sri Lanka for managing LPHS, guided by expert opinion. This study will try to determine the effect of moderate-dose or high-dose steroid combination with plasmapheresis compared with plasmapheresis alone as a treatment for LPHS.

Methods and analysis

This open-label, randomised controlled trial will enrol LPHS patients from four selected tertiary care hospitals in Sri Lanka. Participants will be randomly assigned to three groups in a 1:1:1 ratio to receive plasmapheresis alone, plasmapheresis with 150 mg/day methylprednisolone for 3 days or plasmapheresis with 1000 mg/day methylprednisolone for 3 days. Recruitment will consist of 28 participants to each arm of the trial. Diagnosis will be confirmed by Leptospira-specific PCR and microscopic agglutination test. Analysis will follow intention-to-treat principles with adjustment for multiple comparisons. Primary outcome is all-cause mortality at 28 days. Secondary outcomes include hospital-acquired infections, duration of respiratory support, duration of intensive care unit stay, duration of hospital stay and side effects of steroid therapy.

Ethics and dissemination

The ethical approval was obtained from the Research Ethics Committee of the University of Sri Jayewardenepura (ERC 20/24) and the trial has been registered in the Sri Lanka Clinical Trials Registry (SLCTR/2024/037). Trial is approved by the National Medicines Regulatory Authority in Sri Lanka (NMRA/CTRD/PA4/54). Study findings will be disseminated through peer-reviewed publications and conference presentations.

Trial registration number

Sri Lanka Clinical Trials Registry, Number SLCTR/2024/037.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This randomised trial explores the effect of two steroid regimens as an adjunct to plasmapheresis treatment compared with plasmapheresis alone in leptospirosis pulmonary haemorrhage syndrome (LPHS).

• The study will include patients with a robust confirmation of leptospirosis by PCR and microscopic agglutination test to ensure avoidance of leptospirosis mimics which will enhance the internal validity.

• The effect size assumed in sample size may be optimistic which will be adjusted during an interim analysis.

• Heterogeneity across centres will be minimised by the use of standard treatment guides across all four centres managing LPHS patients in the trial.

• Open-label design introduces potential performance and detection bias, although objective primary outcomes (mortality) and standardised secondary outcome definitions mitigate this limitation.

Introduction

Leptospirosis is a zoonotic disease of global concern caused by pathogenic spirochetes of the genus Leptospira. Humans are infected by exposure to contaminated water, and the organism enters the bloodstream via intact mucosa or abraded skin.^{1 2} It is a gram-negative spirochaete, and the genus Leptospira has approximately 22 species encompassing 300 serovars belonging to pathogenic, saprophytic and intermediate groups.³ Leptospirosis has approximately one million cases reported annually throughout the world, with significant mortality with increased utility of healthcare resources.^{4 5} Leptospirosis is a major cause of morbidity and mortality in Sri Lanka.⁶ Leptospirosis pulmonary haemorrhage is a significant cause of death among leptospirosis patients.

The pathogenesis of leptospirosis pulmonary haemorrhage syndrome (LPHS) is still under investigation. It has been hypothesised that direct toxic injury to pulmonary pneumocytes and blood vessels, as well as immune-mediated lung injury leading to pulmonary haemorrhage, plays a role. Treatment of this lethal complication varies across regions and there is no consensus on the best treatment modality for LPHS at present. Lack of robust clinical trials has hampered development of evidence-based management protocols globally. Countries affected by leptospirosis adhere to treatment modalities based on expert opinion and evidence generated from observational studies. At present, plasmapheresis is routinely performed for patients with LPHS in Sri Lanka based on the benefit demonstrated in observational studies.⁷

Based on the understanding of LPHS pathogenesis, case reports of clinical improvement with steroids (pulse methylprednisolone, dexamethasone with or without oral prednisolone) have been published. The only randomised clinical trial comparing steroid treatment (intravenous dexamethasone 200 mg for 3 days followed by oral prednisolone 1 mg/kg for 4 days) versus standard care did not show a benefit in patients treated in Thailand.⁸ This trial enrolled patients with suspected leptospirosis, who, according to the treating physician, were likely to have pulmonary involvement. However, the diagnosis of pulmonary haemorrhage was not robust in this trial; despite many observational studies reporting chest radiograph changes in 100% of patients with pulmonary haemorrhage, 5 out of 68 patients (7.3%) had normal chest radiographs. The national guidelines for leptospirosis management in Sri Lanka, published in 2016,⁹ suggest the use of intravenous methylprednisolone 1000 mg for 3 days, followed by oral prednisolone, but acknowledge that the evidence is of low quality. In contemporary practice, plasmapheresis is routinely performed for LPHS patients when available, while corticosteroid use varies substantially—some practitioners adhere to guideline-recommended high-dose regimens, others employ modified lower doses and some avoid steroids entirely due to concerns about immunosuppression and secondary infection. There is no clear consensus on the use of steroids as an adjunct to plasmapheresis. However, no trials have been conducted to assess the efficacy of steroids in patients treated with plasmapheresis, creating a significant gap.

Previous observational data from our group¹⁰ suggest that steroid use combined with plasmapheresis does not improve survival compared with plasmapheresis alone. However, steroid treatment increased the risk of secondary infections. Therefore, more robust evidence is needed to confirm the actual benefit of steroids in treating LPHS as an adjunct to plasmapheresis. The dose of methylprednisolone used has not been tested in previous studies. COVID-19, which can cause pulmonary involvement and hypoxia, is treated with intravenous dexamethasone 6 mg (equivalent to 32 mg of methylprednisolone) based on clinical trials. Higher doses have not demonstrated better outcomes. Lower doses of methylprednisolone have not been evaluated in LPHS. We propose to evaluate whether intravenous methylprednisolone 150 mg is equal in efficacy in patients receiving plasmapheresis compared with intravenous methylprednisolone 1000 mg. Such dosing regimens have not been previously assessed in clinical trials.

The primary research question of the study is whether the addition of steroids to plasmapheresis has an effect on the 28-day mortality of patients with LPHS compared with plasmapheresis alone. Additionally, we aim to determine whether a lower dose of steroids at 150 mg, compared with 1000 mg, provides equal efficacy. We hope to examine the effect of interventions on the duration of respiratory support needed for LPHS, as well as intensive care unit (ICU) and hospital stay lengths. Furthermore, we will investigate the side effects of steroid treatment, such as hyperglycaemia, hyponatraemia or other electrolyte abnormalities, in patients receiving the two different doses of intravenous methylprednisolone. The proposed exploratory phase II pragmatic trial intends to generate preliminary efficacy estimates, safety data and feasibility data for future larger multicentre trials.

Aim and study objectives

The aim of our study is to evaluate whether the addition of intravenous methylprednisolone (150 mg or 1000 mg daily for 3 days) to plasmapheresis reduces 28-day mortality in patients with LPHS compared with plasmapheresis alone. Specific objectives of the study are as follows:

1. To compare the 28-day mortality of combined intravenous methylprednisolone (150 mg and 1000 mg) and plasmapheresis therapy compared with plasmapheresis alone in the treatment of LPHS.

2. To compare the development of hospital-acquired infections (ventilator-associated pneumonia, urosepsis, indwelling catheter infections, cannula site infections) in moderate-dose or high-dose intravenous methylprednisolone and plasmapheresis combined therapy versus plasmapheresis alone for LPHS at 28 days.

3. To compare the duration of respiratory support (length of mechanical ventilation, duration of high-flow nasal oxygen (HFNO)/non-invasive ventilation (NIV) therapy) required in moderate-dose or high-dose intravenous methylprednisolone and plasmapheresis combined therapy versus plasmapheresis alone in LPHS.

4. To compare the duration of ICU stay in patients treated with moderate-dose or high-dose intravenous methylprednisolone and plasmapheresis combined therapy versus plasmapheresis alone in LPHS.

5. To compare the duration of hospital stay in patients treated with moderate-dose or high-dose intravenous methylprednisolone and plasmapheresis combined therapy versus plasmapheresis alone in LPHS.

6. To describe the side effects of steroid therapy (hyperglycaemia, hyponatraemia, symptomatic peptic ulcer disease) in moderate-dose and high-dose intravenous methylprednisolone-treated groups in LPHS.

Trial design

This is an open-label, randomised controlled trial designed to assess the effect of moderate-dose and high-dose intravenous methylprednisolone treatment as an adjunct to plasmapheresis. The participants will be randomised in a 1:1:1 ratio to three arms: (1) plasmapheresis with moderate-dose methylprednisolone (150 mg daily for 3 days), (2) plasmapheresis with high-dose methylprednisolone (1000 mg daily for 3 days) and (3) plasmapheresis alone (control). The randomisation is generated through a coding system. The trial adheres to the Standard Protocol Items: Recommendations for Interventional Trials guidelines and will be reported in accordance with Consolidated Standards of Reporting Trials statements.

Methods and study design

Trial setting: the study will recruit patients from the Colombo South Teaching Hospital, the National Hospital of Sri Lanka, National Hospital Galle and Colombo North Teaching Hospital. Patients will be recruited from the medical wards and ICUs at the above hospitals. The study will be conducted over a period of 2 years starting from the day of first enrolment with an additional 12 months for follow-up completion and data analysis. The first patient was recruited on 28 October 2025, and the recruitment is planned to be completed by October 2027.

Study subjects

Patients suspected of leptospirosis will be recruited from the wards or ICUs. Leptospira PCR and microscopic agglutination test (MAT) will be performed on whole blood and serum samples to confirm the diagnosis. As patient recruitment occurs early, before the repeat MAT result is available, Leptospira investigation-negative patients will be included in the intention-to-treat analysis but excluded during per-protocol analysis. Patients will be followed up for 28 days after admission.

Eligibility criteria

Inclusion criteria include suspected leptospirosis patients who develop LPHS admitted to selected medical wards or ICUs at the study sites will be included. Eligible patients will be recruited within 24 hours of development of LPHS. Patients less than 18 years of age or pregnant, patients on long-term glucocorticoids, patients with active co-infections at recruitment (eg, dengue, sepsis, tuberculosis, systemic fungal infections), patients with underlying bleeding disorders and patients with pre-existing lung diseases interfering with chest radiograph interpretation (eg, interstitial lung disease, bronchiectasis) will be excluded (figure 1).

Figure 1. Recruitment flow chart. The flow diagram represents the initial recruitment, randomisation and the intervention groups of the randomised controlled trial. LPHS, leptospirosis pulmonary haemorrhage syndrome.

Case definitions

The criteria for suspected leptospirosis, confirmed leptospirosis and diagnosis of LPHS are given below.

• Suspected leptospirosis is identified by the treating clinician/team. An acute febrile illness associated with neutrophil leukocytosis, thrombocytopenia and clinical symptoms defined in the national guideline⁹ is used to identify a ‘suspected case’ of leptospirosis.

• Confirmed leptospirosis is defined as a suspected case of leptospirosis with a confirmatory test of leptospirosis (positive Leptospira-specific PCR in blood and/or positive MAT).

• LPHS is defined as a suspected leptospirosis patient with one or more of the following: presence of haemoptysis, arterial hypoxaemia, haemoglobin reduction (10% or more reduction from baseline), with diffuse alveolar shadows in the chest radiograph, suggestive of pulmonary haemorrhage, without an alternative explanation other than leptospirosis.¹⁰ High-resolution chest radiograph will be used if the chest radiograph is inconclusive and reported by a consultant radiologist for accuracy.

  Previous studies report that all LPHS patients have radiological abnormalities.¹¹ Therefore, abnormal chest radiography is considered a mandatory criterion for the diagnosis of LPHS in the study to increase the robustness of the diagnosis.

Subject recruitment

Method of recruitment

All study participants will be recruited to the study by a trained research team member from the medical ward or ICUs of the study sites once identified as a suspected leptospirosis patient with LPHS.

Potential participants will be approached by the trained research assistant (independent of the treating team) to explain the study and give the written information sheet in their respective language. If the participant agrees to be enrolled in the study, a consent form will be signed (online supplemental file 1). Where participants lack decision-making capacity at enrolment, informed consent will be obtained from a legally authorised representative or immediate relative. If the participant subsequently regains decision-making capacity, re-consent will be sought whenever feasible.

Randomisation

All recruited LPHS patients at the time of diagnosis of pulmonary haemorrhage will be randomised to one of the three arms (control group, moderate-dose steroids or high-dose steroids) in each centre. Randomisation will be with equal numbers for each treatment arm. The trial is an open-label parallel trial; therefore, the treatment given will be known to the treating medical team and the trial team. Cross-over to other arms will not be considered. The final analysis will be conducted blinded to treatment allocation by the statistician.

A balanced design will be used, meaning that each participant has an equal chance of being assigned to any of the three treatment arms (ratio 1:1:1). A centralised randomisation system will be used. At each participating centre, when a participant is deemed eligible and provides informed consent, the site contacts the central randomisation unit. The randomisation unit then assigns the participant to one of the three treatment arms using randomisation codes prepared by an independent statistician.^{12 13} Treatment records for each participant will be kept in a secure and auditable manner. All centres will implement monitoring procedures to ensure adherence to the randomisation protocol. Adaptations to the randomisation process will be considered based on any emerging issues or challenges during the trial, such as changes in recruitment rates or unexpected dropouts.

Interventions

Study participants will be randomly allocated to the following treatment groups.

• Control arm: plasmapheresis alone.

• Intervention 1: plasmapheresis and intravenous methylprednisolone 150 mg daily for 3 days.

• Intervention 2: plasmapheresis and intravenous methylprednisolone 1000 mg daily for 3 days.

The trial will be using methylprednisolone at a dose of 1000 mg or 150 mg delivered intravenously from a cannula placed in the patient by a trained nurse in the ICU. Use of 150 mg intravenous methylprednisolone is considered a moderate dose, and it will be used to see the efficacy and side effects against high-dose 1000 mg methylprednisolone. The duration of treatment is 3 days, as used in previous studies of steroid treatment for LPHS.⁸ Plasmapheresis will be performed by exchanging one plasma volume with fresh frozen plasma, with a percentage replacement of 90–95%. The patient’s plasma volume is calculated automatically by the machine using the patient’s height, weight and haematocrit.

Other treatment

All patients will receive standard of care at the discretion of the treating physician and intensivist as described below.

• Supplemental oxygen, high-flow nasal oxygen therapy, non-invasive ventilation or invasive mechanical ventilation as appropriate.

• Haemodialysis or Continuous Renal Replacement Therapy (CRRT) for acute kidney injury.

  Renal replacement therapy decisions will follow institutional critical care protocols based on Kidney Disease: Improving Global Outcomes (KDIGO) -guided indications, including refractory hyperkalaemia, severe metabolic acidosis, fluid overload and uremic complications. The modality used (intermittent haemodialysis or CRRT) and timing of initiation will be documented for all participants.

• Standard antibiotics—intravenous ceftriaxone 1 g two times per day for 7 days.

• Platelet transfusions in the presence of thrombocytopenia <50 000/mm³, fresh frozen plasma transfusions for coagulopathy (INR >1.5 or identified deficiency by Rotational Thromboelastometry).

Participants randomised to control or intervention arms may not receive open-label systemic corticosteroids for any indication during the 3-day intervention period, except as rescue therapy for life-threatening refractory shock at treating clinician discretion (such cases will be analysed as intention-to-treat with sensitivity analyses) and such deviations will be documented.

Procedures

A total blood sample of 10 mL at the time of recruitment will be collected from the patient following consent for Leptospira PCR and leptospirosis MAT. MAT will be performed on the acute sample at recruitment. If this sample does not fulfil diagnostic criteria for leptospirosis (titre >1:400), MAT will be performed in the convalescent sample at 10–14 days to identify seroconversion or a fourfold rise of titre. A data collection form will be used to gather information on basic demographic data, clinical symptoms during the illness and laboratory findings. The research assistant will complete the study instrument.

Leptospirosis diagnostic tests

MAT will be performed using a panel of 15 Leptospira serogroup strains. A MAT titre of ≥1:400, the cut-off value recommended by the WHO, will be considered as positive. A fourfold rise in titre in the convalescent sample or seroconversion will also be considered a positive test.

Leptospira-specific quantitative PCR

The QIAamp DNA Blood Mini Kit (QIAGEN, Germany) will be used for rapid extraction of Leptospira DNA from 200 µL of whole blood collected into ethylenediaminetetraacetic acid (EDTA) collection tubes according to the manufacturer’s instructions. Identification of Leptospira DNA will be carried out using a real-time qPCR assay with LipL32 primers (IDT, USA), which will amplify the LipL32 gene of pathogenic Leptospira species.

Outcomes

The primary outcome of the study is all-cause 28-day mortality. Secondary outcomes of the study include the development of hospital-acquired infections (ventilator-associated pneumonia, urosepsis, indwelling catheter infections, cannula site infections), the duration of respiratory support, including the length of mechanical ventilation and the duration of HFNO or NIV therapy, as well as the length of ICU stay, and total hospital stay. The study will also assess the side effects of steroid therapy, specifically the occurrence of hyperglycaemia, hypertension, hyponatraemia and peptic ulcer disease.

Sample size calculation

Sample size was calculated considering the primary outcome of mortality. A reduction in mortality from 60% to 20% in either plasmapheresis and intravenous methylprednisolone 150 mg daily for 3 days (intervention 1) or plasmapheresis and intravenous methylprednisolone 1000 mg daily for 3 days (intervention 2) compared with plasmapheresis alone (control group) with 95% confidence and 80% power.^{14 15}

Mortality in LPHS is reported to be 40–60%^{8 9} in the national guidelines for management of leptospirosis in Sri Lanka and previous literature. A reduction in mortality from 60% to 20% was considered the effect size based on previous data and the published randomized controlled trial.⁸

According to the calculation, the minimum sample size required for each arm would be 28 subjects, considering a 5% dropout rate. The patients will be recruited from all four study sites simultaneously. One formal interim analysis after 50% of participants (n=42) have completed 28-day follow-up will be conducted using O’Brien-Fleming stopping boundaries (α=0.005 at interim) to find the power of the sample size during the study and to consider sample size expansion.

Data collection and follow-up

A trained research assistant will be collecting the information on basic demographic data, clinical symptoms during the illness and laboratory findings of the trial participants using case report forms. All recruited patients will be followed up until 28 days after admission to obtain data and necessary blood samples. Data will be collected using a data extraction sheet. Data on demographic details, clinical manifestations, vital signs, diagnosis of LPHS and investigation results, including radiological findings, will be collected by the trained research assistant. Length of hospital stay, need for intubation, length of mechanical ventilation/respiratory support, mortality, occurrence of secondary infections and selected side effects of steroid treatment will be documented.

Data management

Access to data and samples

Only the investigators included in the study will have access to data, and such information will be dealt with extreme confidentiality. Samples will be stored at the Department of Medicine, Faculty of Medical Sciences, at University of Sri Jayewardenepura. The blood samples will be stored for 10 years and discarded by the clinical waste disposal method stipulated in the standard operating procedure of the department. Data will be stored for 10 years with the principal investigator. Data will be recorded and kept with the primary investigator mentioned in the study proposal and will not be available for access by anyone outside the research group.

Participant confidentiality

Participant data will not be saved by their name or in any other identifiable manner. The patient ID (bed head ticket no) will be used for all identification purposes of the patient information. Data will not be divulged to anyone outside the research group.

Statistics and analysis

The efficacy of the treatments will be analysed on an intention-to-treat basis whereby all randomised participants will be analysed according to their allocated treatment arm irrespective of protocol deviations. A secondary per-protocol analysis will additionally be performed as an exploratory sensitivity analysis. The primary outcome measure will be the mortality rate at 28 days. Primary outcome will be analysed with pairwise comparisons between (1) low-dose methylprednisolone with plasmapheresis versus plasmapheresis alone and (2) high-dose methylprednisolone with plasmapheresis versus plasmapheresis alone. Exploratory comparisons between the two corticosteroid regimens will additionally be conducted primarily for safety and hypothesis-generating purposes. Intention-to-treat analysis will be conducted to explore whether treatment effects vary across different subgroups (eg, age, gender, disease severity). Data management and analyses will be performed blinded to treatment allocation.

Statistical analysis will be performed using SPSS V.22.0. Baseline statistics of the three arms will be described using central tendency measures. Baseline differences between the three groups will be assessed by either univariate analyses of variance or χ² tests. The pairwise comparisons will be done to determine which specific treatment groups differ significantly using appropriate statistical tests (eg, t-tests, Analysis of Covariance). Adjustments will be made to control the overall type I error rate when using multiple tests. Secondary outcomes will be compared between each treatment group, and regression analysis will be used to determine independent risk factors for death/relevant outcomes where necessary. If non-normally distributed, generalised linear models with appropriate distributions will be used. The single interim analysis at 50% enrolment will use O’Brien-Fleming boundaries. An independent review board will receive unblinded comparative data and make recommendations to the Trial Committee regarding continuation, modification or termination. It may recommend early stopping for (1) overwhelming efficacy in either steroid arm (mortality reduction exceeding prespecified threshold), (2) futility (conditional power <20%) or (3) safety concerns (significant excess of serious adverse events).

Monitoring

All adverse effects reported in the treatment team will be recorded in the data collection sheet and informed to the Research Ethics Committee in the progress reviews every 6 months. An independent safety monitoring board will be appointed to monitor adverse events related to the trial. Review members will consist of physicians not involved in the study including a consultant physician, consultant intensivist and a consultant paediatrician.

Ethics and dissemination

Only individuals who agree to participate in the study after reading the information sheet, which is provided in their native language, are eligible. If a participant is unable to read, their legal representative or accompanying person must read the information sheet to them and sign it on their behalf. Participants have the right to withdraw from the study at any time, as decided by the patient themselves or their immediate relative. Such participants will be included as dropouts and analysed in the intention-to-treat analysis. Patients and the public were not involved in the study design and will not be involved in the conduct of the study. Study findings will be disseminated through peer-reviewed publications and conference presentations.

This study was approved by the Research Ethics Committee of the University of Sri Jayewardenepura (ERC 20/24). This study was registered in the Sri Lanka Clinical Trials Registry (SLCTR) bearing number SLCTR/2024/037 on 21 November 2024 (URL: https://slctr.lk/trials/slctr-2024-037). The trial is approved by the National Medicines Regulatory Authority in Sri Lanka (NMRA) (NMRA/CTRD/PA4/54). Institutional Ethics Review Committee approvals at the study sites were also obtained from the National Hospital of Sri Lanka, Colombo South Teaching Hospital, National Hospital Galle and Colombo North Teaching Hospital.

Discussion

This trial addresses a critical evidence gap in the management of LPHS, a condition with persistently high mortality and limited therapeutic options. The addition of corticosteroids to plasmapheresis represents a logical extension of current understanding of LPHS pathogenesis, which emphasises immune-mediated alveolar injury.¹⁶ However, the absence of randomised data leaves clinicians navigating between (a) theoretical benefits of intensified immunosuppression and (b) concerns about secondary infections. As such, the current trial will provide useful clinical information for managing physicians.

Strengths

This trial has several methodological strengths. First, the multicentre design enhances generalisability across Sri Lanka’s diverse geographic regions and healthcare settings. Second, robust diagnostic confirmation using both PCR and paired MAT ensures that analysis excludes leptospirosis mimics, a common challenge in observational studies. Third, the dose-comparative design generates novel evidence regarding the minimum effective dose—particularly important if steroids prove beneficial but high-dose regimens carry excessive toxicity. Fourth, comprehensive secondary outcomes capture not only mortality but also resource utilisation and safety, informing cost-effectiveness and risk-benefit assessments.

Limitations

Several limitations warrant acknowledgement. The open-label design introduces potential bias, particularly for subjective outcomes. However, the primary outcome (mortality) is objective, and we have implemented blinding for outcome assessors where feasible. The sample size, while adequate for the primary comparison, may be insufficient for definitive subgroup analyses or detection of modest differences in secondary outcomes. The ambitious effect size assumption (60% to 20% mortality reduction) means the trial is powered to detect a large effect; smaller but clinically meaningful benefits may be missed. However, if any small clinically meaningful effects are observed, this exploratory trial will inform conduct of further larger multicentre trials in the future. If the interim analysis suggests underpowering, we will in addition consider sample size expansion.

Major limitation is the different treatment modalities practised globally. Although plasmapheresis is a standard practice in Sri Lanka, it is not widely adopted in other countries. The current question of adjunctive steroids in the background plasmapheresis treatment could be difficult to interpret for clinical use in countries where such modalities are not used. Plasmapheresis availability varies globally and our findings will apply primarily to centres with this capability.

The exclusive focus on Sri Lankan patients may limit generalisability to other regions with different leptospirosis serovars, host genetics or healthcare infrastructure. However, the immune-mediated pathogenesis of LPHS appears conserved across settings, suggesting biological plausibility for similar treatment effects elsewhere.

Implications

If steroids demonstrate benefit, this trial will provide evidence to support and optimise a readily available, low-cost intervention applicable in resource-limited settings where leptospirosis burden is highest.

Data availability statement

No data are available.