Clinical characterization of Lassa fever: A systematic review of clinical reports and research to inform clinical trial design.

BACKGROUND: Research is urgently needed to reduce the morbidity and mortality of Lassa fever (LF), including clinical trials to test new therapies and to verify the efficacy and safety of the only current treatment recommendation, ribavirin, which has a weak clinical evidence base. To help establish a basis for the development of an adaptable, standardised clinical trial methodology, we conducted a systematic review to identify the clinical characteristics and outcomes of LF and describe how LF has historically been defined and assessed in the scientific literature.
METHODOLOGY: Primary clinical studies and reports of patients with suspected and confirmed diagnosis of LF published in the peer-reviewed literature before 15 April 2021 were included. Publications were selected following a two-stage screening of abstracts, then full-texts, by two independent reviewers at each stage. Data were extracted, verified, and summarised using descriptive statistics.
RESULTS: 147 publications were included, primarily case reports (36%), case series (28%), and cohort studies (20%); only 2 quasi-randomised studies (1%) were found. Data are mostly from Nigeria (52% of individuals, 41% of publications) and Sierra Leone (42% of individuals, 31% of publications). The results corroborate the World Health Organisation characterisation of LF presentation. However, a broader spectrum of presenting symptoms is evident, such as gastrointestinal illness and other nervous system and musculoskeletal disorders that are not commonly included as indicators of LF. The overall case fatality ratio was 30% in laboratory-confirmed cases (1896/6373 reported in 109 publications).
CONCLUSION: Systematic review is an important tool in the clinical characterisation of diseases with limited publications. The results herein provide a more complete understanding of the spectrum of disease which is relevant to clinical trial design. This review demonstrates the need for coordination across the LF research community to generate harmonised research methods that can contribute to building a strong evidence base for new treatments and foster confidence in their integration into clinical care.

Introduction

Lassa Fever (LF) is an acute viral haemorrhagic disease caused by the Lassa virus, which is endemic to parts of West Africa, including Nigeria, Sierra Leone, Guinea and Liberia. Transmission to humans usually occurs through contact with excreta of infected rodents (primarily the Mastomys rat), [1,2] consumption or handling of contaminated food or household items (mostly involving women and children), or through direct contact with the bodily fluid of an infected person (typically healthcare workers) [1,3].

LF is a seasonal disease that is estimated to cause 100,000 to 300,000 new cases and 5,000 deaths each year. [4] The reported case fatality rate (CFR) is approximately 30% in patients who present to health care settings [5]–although this figure is lower (12%) in the most recent large cohort study taking place in a research setting. [6] During the peak season it is possible for large outbreaks to occur. Nigeria has experienced increases of confirmed cases and deaths every year since 2017, although this may be attributed to heightened clinical awareness and improvements in diagnostic capacity. [7-10]

Healthcare workers and pregnant women are considered to be at significant risk of severe Lassa fever outcomes. [11] High prevalence of LF has been identified in healthcare workers [5] which is thought to be a result of low levels of clinical suspicion of LF and an inadequate supply of quality protective equipment, making adherence to infection prevention and control (IPC) measures challenging. [12,13] Pregnant women are three-times more likely to have a fatal outcome than non-pregnant adults. [14]

No drug has so far received regulatory approval for treating LF. Ribavirin, in conjunction with supportive care, is currently used as the primary treatment for LF and has been incorporated into national and international treatment guidelines. [2,15] Ribavirin is on the World Health Organisation (WHO) list of essential medicines for treating viral haemorrhagic fevers. [16] However, this treatment recommendation has a small evidence base, as only a single clinical trial has been conducted to evaluate its effectiveness. [17,18] Further evidence in the form of clinical trials is required both to confirm the efficacy and safety of ribavirin and to test new therapies.

Multiple reasons conspire to make LF a neglected, difficult-to-study infectious disease of poverty. The narrow geographical spread of LF, which translates into a relatively limited number of patients who can be treated and studied in the few available specialised healthcare facilities, may contribute to low commercial interest for pharmaceutical companies. Consequently, a significant challenge to building the clinical evidence base is the lack of a methodological framework in which LF clinical trials can be reliably conducted in a consistent and comparable manner.

To establish a basis for the development of a standardised clinical trial methodology, we conducted a systematic review to identify the clinical characteristics and outcomes of LF, and to understand how LF has historically been defined and assessed in the scientific literature. The clinical characterisation enabled by this review is examined in combination with the diagnostics and demographics across the literature to expose the full spectrum of the disease that should be considered in optimising clinical trial methods.

This approach has previously been used to establish a foundation for the development of harmonised clinical trial methodologies, and specifically for the development of Core Outcome Sets, for other infectious diseases. For example, two systematic reviews of cutaneous leishmaniasis interventions conducted by Gonzalez and colleagues (2008 & 2009) [19,20] formed the knowledge-base for standardised trial methodology proposed by Olliaro and colleagues (2013)[21], and a review of tuberculosis meningitis research informed the definition of standardised diagnostic criteria, [22] data collection and outcomes for future clinical trials.[23]

Methodology

An initial search of Epistemonikos and Prospero was conducted to understand if any high-quality reviews were available or in progress that covered portions of the planned review.

For the main search, the following databases and clinical trial registries were searched for clinical studies: African Journals Online, Cochrane Central Register of Controlled Trials (CENTRAL), Embase, Global Health (Ovid), Global Index Medicus, PubMed/MEDLINE, clinicaltrials.gov, ISRCTN, Pan African Clinical Trials Registry, and WHO International Clinical Trials Registry. Cochrane Database of Systematic Reviews, Cochrane Clinical Answers, and COMET databases were also searched (). Studies of LF published before 15 April 2021 were included with no language restrictions applied. We registered this review in PROSPERO, the international prospective register of systematic reviews of the University of York and the National Institute for Health Research, under protocol number CRD42020220365. [24]

We included only studies reporting primary results of patients with a laboratory or clinically confirmed diagnosis of LF (including all diagnostic methods and studies that report patients with sequelae from a previous infection) and describing LF clinical features and/or treatments.

All identified studies underwent screening for inclusion/exclusion in two stages: first by review of titles and abstracts, then by review of the full text manuscripts. Screening was conducted in Rayyan, [25] aided by pre-defined inclusion/exclusion key words. At both stages of the review, decisions on the inclusion or exclusion of each study required the agreement of two unique reviewers assessing the reference or manuscript independently. Each reviewer’s decisions were blinded to the second reviewer. When a dataset was fully screened by both reviewers, decisions were unblinded and conflicts were resolved by discussion between the two reviewers, or with a third reviewer when required, to agree on inclusion or exclusion.

Data were extracted from the selected manuscripts and entered into a Research Electronic Data Capture (REDCap) database using a predefined variable dictionary available in . [26] A second reviewer verified all extracted data and resolved any discrepant data by reviewing the manuscript with a third reviewer. During full-text screening, the most frequently reported signs and symptoms of interest were identified by the review team. For each publication, the presence or absence of information on each of these 19 pre-defined signs and symptoms was recorded where available. Several other signs and symptoms of LF were also reported. Data on these characteristics were extracted and assessed for clinical significance by clinicians in the study team. Signs and symptoms considered to be clinically significant are reported alongside those that were pre-identified. All other reported signs and symptoms can be found in .

For the sections in this systematic review that report data relating to case fatality, signs and symptoms of LF, data are presented only for studies that include populations with 100% laboratory confirmation (N = 122)–all other sections in this review report data for all publications (N = 147) regardless of the proportion of the study population with laboratory confirmation of LF.

The prevalence of signs and symptoms is reported based on the number of patients for whom there was evidence of assessment of each sign or symptom in the publication (number of patients with sign or symptom reported/over number of patients assessed). Results relating to bleeding site are reported as a percentage of publications that include both location of bleeding site and number (or proportion) of individuals exhibiting bleeding at specified sites.

The minimum and maximum time in days were extracted for data relating to time from symptom onset to presentation and is reported alongside the median, interquartile range (IQR) and range. For data relating to time from presentation to death, the median of reported mean times in days, the IQR and the range are reported. Aspartate aminotransferase (AST) is reported categorically as <150 IU/L and ≥150 IU/L to understand liver enzyme levels in the context of the evidence base that has informed current treatment guidelines. [17]

Risk of bias assessments were conducted using the Joanna Briggs Institute (JBI) critical appraisal tools due to the compatibility of these tools with the variety of study types in this review.[27,28] Tools were applied to each study based on criteria outlined in the JBI Manual for Evidence Synthesis. [29] Diagnostic methods were evaluated for confidence of acute LF diagnosis based on acceptance in the existing literature and the results of recent reviews. [30,31]

Results

Search results

In total, 4,794 publications were identified in the literature search. After removing duplicates, 2,704 titles and abstracts were screened and 195 full-text publications were assessed for eligibility, resulting in the inclusion of 147 publications in the data synthesis ().

Of the included publications, 53 (36%) were case reports, 41 (28%) were case series, 30 (20%) were cohort studies, 10 (7%) were case-control studies, 11 (7%) were cross sectional studies, and 2 (1%) were quasi-randomised studies ().

The publication years ranged from 1970 to 2021 with increasingly higher numbers of publications being generated after 2010 ().

Risk of bias assessments

Overall, there was a moderate risk of bias across all outcomes evaluated in the risk of bias assessment (). An average of 62% (IQR 44–83%) of study-specific criteria in the JBI Critical Appraisal Checklists were included in the publications. The criteria most relevant to this review were often missing or unclearly reported including: Inclusion criteria clearly defined in 51% of 47 publications, Demographic characteristics described in 34% of 88 publications, Exposure measured reliably in 62% of 109 publications, and Adverse events reported in 53% of 60 publications. These gaps in reporting may result in an incomplete characterisation of LF throughout the clinical course of disease. Though an appropriate statistical analysis was reported in 96% of 79 publications, strategies to deal with confounders were reported in only 19% of 79 publications.

Furthermore, 74 (50%) publications are case reports or case series describing fewer than 6 individuals. The basis for the selection of the enrolled participants is often unclear and may not be representative of the patient population who present to the health centre, meaning generalising the findings of these studies to the wider LF patient population is not feasible.

Population characteristics

In total, data on 8550 individuals were reported in the selected publications (). Most individuals (91%) in the reported population were enrolled in studies conducted in either Sierra Leone or Nigeria. The remainder of the study population were enrolled in studies conducted in West Africa, Europe, Asia and North America ().

Note that reporting details do not enable the delineation of overlapping patient populations reported across multiple publications. Therefore, some patients may have been included in more than one publication.

Of the individuals included in the publications where sex was noted, 3477 individuals (48%) were male and 3839 were female (52%). Sex was not reported for 1234 individuals (14%).

Of the publications in which the age range of patients was reported (N = 129), 48 (37%) included a patient under the age of 16 years. 35 publications (24%) reported the inclusion of at least one pregnant participant.

Eligibility criteria

68 publications (46%) documented eligibility criteria (). Of these, all 68 (100%) reported inclusion criteria and 5 (7%) reported exclusion criteria.

Within the inclusion criteria, 20 publications (29%) specified signs and symptoms that must be present for inclusion, 17 (25%) specified fever should be present upon enrolment. 27 publications (40%) required laboratory confirmation of LF either alone or in conjunction with clinical symptoms. 26 publications (38%) required a confirmation or diagnosis of LF without specifying a method of diagnosis.

Exclusion criteria were based on geographic location in 2 publications (3%). Hearing loss, co-infection with Marburg virus, contacts of an index LF case, patients with an incomplete data set, and LF patients managed on an out-patient basis were each an exclusion criteria in 1 publication (1%).

Outcome measures

Six publications (4%) defined a primary endpoint or outcome measure, 5 (83%) of which were cohort studies and 1 (17%) of which was a quasi-randomised trial. Five (83%) publications included mortality as the primary outcome measure and one (17%) included diagnosis of acute kidney injury (AKI).

Three publications (2%) reported at least one secondary endpoint or outcome measure. Viral load throughout treatment was an outcome measure in 2 publications (66%). Aspartate aminotransferase (AST) throughout treatment, live birth, all-cause in-hospital fatality, frequency of acute kidney dysfunction, prognosis of AKI and AKF in terms of estimated glomerular filtration rate (eGFR) at the end of follow-up and time to hospital discharge were each listed as an outcome measure in 1 publication.

The period of follow-up was specified in 97 publications (66%). The median follow-up time was 21 days (IQR 9–67) with a range of 1–10,950 days.

Method of case confirmation

Laboratory-confirmed diagnosis was the most prevalent method of case confirmation, used either alone or in conjunction clinical diagnosis in 141 publications (96%) and 8422 individuals (99%). Two publications (1%) confirmed LF using clinical diagnosis alone and 4 publications (3%) did not specify a method of case confirmation.

The largest proportion of laboratory diagnoses, for 2348 individuals (27%), was conducted using RT-PCR alone (. Results are shaded according to the level confidence of laboratory methods for the identification of acute LF infection.)

When RT-PCR, viral culture and antigen ELISA were used exclusively or in combination with each other, we determined a high level of confidence of acute LF infection. A moderate level of confidence was determined when serology (IgM ELISA and/or IgG ELISA) was used in combination with RT-PCR, viral culture or antigen ELISA. Any other exclusive or combined use of laboratory methods was determined as a lower confidence of acute LF infection.

Baseline clinical characteristics

At baseline, fever was the most reported symptom, identified in 88% of the individuals in whom it was assessed (1527/1730) (, followed by headache (809/1622 individuals, 50%), vomiting (806/1613, 49%), abdominal pain (660/1581, 42%) and cough (556/1581, 35%).

A smaller number of patients presented with clinically severe or life-threatening signs and symptoms such as shock (12/187, 6%), breathing difficulty (21/310, 7%), and seizure (13/517, 3%). In pregnant women, labour complications were reported in 2/7 (29%).

Overall, musculoskeletal disorders, nervous system disorders and gastrointestinal illnesses were most reported. (

The reported timeframe from symptom onset to presentation ranged from 0–32 days (median reported minimum time: 5 days; median reported maximum time: 8 days). The minimum and maximum times to presentation were similar in Nigeria, Sierra Leone and Liberia, the three endemic countries with >5 studies included.

Post-baseline clinical characteristics

In post-baseline assessment, fever was again the most prevalent symptom, reported for 3067/3300 individuals (93%)–representing a 5% increase in prevalence from baseline.

After fever, headache (2033/3200, 64%), vomiting (1695/3077, 55%) and abdominal pain (1594/3039, 52%) occurred in the highest number of individuals. The prevalence of all signs and symptoms related to the gastrointestinal system increased post-baseline, except nausea.

The number and prevalence of reported severe or life-threatening signs and symptoms also increased post-baseline. Shock had the greatest increase in prevalence from baseline to post-baseline, followed by breathing difficulty and seizure. Respiratory failure and renal failure were reported only at post-baseline timepoints.

1896/6373 (30%) individuals were reported to have died. 109 publications (74%) reported at least one death. 41 publications (31%) reported the time in days from presentation to death and within these publications, the median of the mean times reported per publication from presentation to death was 7.5 days (IQR 3–11) with a range of 0–21 days.

Site of bleeding

Bleeding was reported in 30 (20%) and 53 (36%) publications at baseline and post-baseline, respectively. At baseline, site of bleeding was reported for 168 individuals and at post-baseline site of bleeding was reported for 1088 individuals ().

Haematuria/blood in urine was the most reported bleeding type at both baseline and post-baseline–reported for 23 individuals (14% of reported bleeding sites) at baseline and 131 individuals (12% of reported bleeding sites) at post-baseline. Haematuria is specified as macroscopic haematuria in a third of the reported cases, while no differentiation between macroscopic and microscopic haematuria was made in two thirds of these cases.

Aspartate aminotransferase (AST)

Aspartate aminotransferase (AST) levels were reported at baseline and post-baseline in 29 (20%) and 43 (29%) publications respectively. These publications included 799 (9%) and 1200 (14%) of the populations described in the literature. Detailed, individual patient data were limited. Many publications reported a single AST value from a single patient, others presented a range and mean value for a group of patients. Of the publications that reported AST levels, 10 publications including a total of 19 (2%) individuals reported all patients as having baseline AST levels below 150 IU/L and 15 publications including a total of 459 (57%) of individuals reported AST values ≥150 IU/L for all individuals. AST values reported in other papers showed a wide range of results. ().

Discussion

Despite fifty years of reporting and research, morbidity and mortality caused by LF remain high in patients presenting to health care settings, especially in high-risk populations such as healthcare workers and pregnant women. [11,14] A historic lack of investment in LF clinical research and LF drug development [179] has resulted in the limited anthology of case reports and observational study reports presented in this review, but no well-conducted randomised trials of current or new therapeutics.

Attention and investment in LF have increased in recent years with the launch of the World Health Organisation (WHO) Research & Development Blueprint [180] and the Coalition for Epidemic Preparedness Initiatives (CEPI) ‘Enable’ project. [181] Since 2018, LF is also included in the US Food and Drug Administration (FDA) list of “infectious disease for which there is no significant market in developed nations and that disproportionately affects poor and marginalized populations”, which entitles the developer of an LF drug successfully registered with the US FDA to a priority review voucher (PRV). [182] These developments will hopefully promote new additions to the sparse evidence presented herein.

Challenges and limitations of the current literature

Both the context and the presentation of LF present challenges to timely diagnosis and treatment, which are reflected in the limitations of the literature included in this review. (186–189) Resources, communications, and awareness of LF are reported to be increasing over time in the most affected countries, [183,184] though challenges to the effectiveness of these interventions are frequently met. [185,186] Differences in LF lineages circulating among the reported populations can also not be considered in this review as this information is not included in the clinical reports nor is clinical information detailed in publications on strain identification [187-189].

Due to these constraints, the limited quality of the literature, and the risk of bias demonstrated in the analysis, our report is strictly descriptive of LF clinical features and cannot inform the impact of LF treatment.

The purpose of this review is to ensure that future research, including the development of standardised data variables and outcomes for LF clinical trials, reflects the entirety of the available evidence base. Additionally, we aim to improve the uniformity of data collection and reporting in future LF research to support efficiency and comparison across studies.

Case definition

The WHO estimates that 80% of cases are asymptomatic. It further defines common symptoms to include gradual onset of fever, malaise and general weakness; after a few days, extending to headache, sore throat, muscle pain, chest pain, nausea, vomiting, diarrhoea, cough, and abdominal pain; as well as possible bleeding, neck/facial swelling and shock in severe cases. [190] The Nigeria Centre for Disease Control (NCDC) suspected case definition is similar, but lists a more restricted set of indicative signs and symptoms of LF. [2] The literature reviewed reveals a much broader spectrum of symptoms than that of both the NCDC and WHO. Expanding the case definition may have advantages e.g., increasing the sensitivity of suspect case identification, the likelihood of rapid access to LF diagnostics, and better protecting healthcare works; however, it may also have disadvantages since these symptoms are non-specific and present in many common febrile illnesses across West Africa, [191,192] the reduced specificity of a case definition that includes them would need to be balanced by the availability of appropriate diagnostics to confirm suspect cases. In the context of clinical research, these issues are tempered by requiring laboratory confirmation for inclusion into the study.

Efforts are currently underway to improve the understanding of regional differences in LF presentation. Data from the CEPI-funded “Enable” project–a large multi-country epidemiological study across five countries in West Africa–aims to summarise the regional prevalence of LF and delineate key differences in its characterisation on a country-by-country basis.[181] These results may inform drivers of diversity in case definition.

Diagnostic variation and confidence

11 different diagnostic approaches are used in the publications included in this review. This variety reflects the history of LF diagnostic development, resource limitations and research on new methods. When multiple tests are used to confirm LF diagnosis, the reported results do not specify which of the tests had a positive result. Furthermore, despite evidence that testing methods, kits and quality assurance protocols have a significant impact on diagnostic quality, [193-195] these details are rarely included in clinical reports and could therefore not be evaluated.

The US Centers for Disease Control and Prevention (CDC) guidelines recommend nucleic acid detection by RT-PCR, antigen detection by ELISA, or serology (IgG or IgM ELISA) for LF diagnosis. [196] The World Health Organisation guidelines concur and further add viral culture as acceptable. [197] In this review, results of studies that confirm LF diagnosis using RT-PCR, antigen ELISA and viral culture exclusively or in combination with each other are determined to have a high level of confidence in the diagnosis. Studies that accept IgM and/or IgG ELISA testing equally with these methods have a reduced confidence on the basis of evidence that detection of IgG and IgM can occur across a wide range of time points ranging from early in an infection and after resolution of acute illness. [198] Investment to improve the accuracy and availability of rapid diagnostics appropriate for LF endemic settings is needed.

Need for more clinical trials

To date, a single clinical trial has been conducted to assess the safety and efficacy of ribavirin, the only recommended treatment for LF. [17] No further clinical trials for LF have been conducted in the subsequent 35 years. The analysis of this trial includes only a fraction of the cohort who received ribavirin and reports only those with an AST >150 IU/L. Reassessment of the results of this trial, and a subsequent report released under the Freedom of Information Act 2000, [199] identified harmful effects of using ribavirin to treat LF in patients with AST <150 IU/L. [18] Despite this significant limitation, ribavirin has been incorporated in to LF treatment guidelines without reference to AST-dependent dosing. [2]

This review cannot contribute much to knowledge of AST levels in LF, as they are minimally reported in the literature (15% of patients in this review). Where results are available, they are not sufficiently detailed to understand the distribution of AST values, possibly due to limited laboratory capacity. [200] In the reported data, publications including 459 (57%) patients reported all patients as having baseline AST levels ≥150 IU/L, above the limit of where harm from ribavirin has been identified. However, publications including only 284 (22%) patients reported all patients having AST levels ≥150 IU/L throughout admission. The potential for harm when AST levels are <150 IU/L after the start of ribavirin treatment has not been examined. Systematic assessment of AST and publication of comprehensive data needs to be undertaken to build understanding of the use of ribavirin in LF.

The shortcomings of ribavirin demonstrate an urgent need for its reassessment and highlight the need for other therapeutic options to be explored–particularly those that can be used in LF patients at heightened risk of death, including pregnant women. [14] There are a number of antivirals that are currently in development or under investigation for LF. [201] With data urgently needed on the safety and efficacy of these new treatment prospects, and in the interest of efficiency, it is vital that clinical trials are conducted and reported in a comparable manner. The spectrum of symptoms and sequalae described in this review provides focus to the targets of therapeutic action and safety considerations.

The importance of standardising data collection and reporting

By collating and describing this history of results, we have shown that understanding of the prevalence of many signs and symptoms of LF is significantly limited by a lack of consistency in reporting of clinical features. There is a clear need for robust reporting of clinical studies and consensus in approaches to clinical characterisation that will allow for comparability of the characterisation of LF across regions and strains of the Lassa virus. The synthesises reported herein serves as a baseline for defining a standardised way to capture and report clinical characteristics across future LF research studies.

Ways forward

Despite the growing number of research outputs on LF, the limited number of cases and the challenges of mounting clinical trials in LF-endemic regions highlight the need for efficient approaches to research. The results of this review have been leveraged to inform the development of standardised clinical trial methodologies with efficient and pragmatic design to address the research priorities above. A consultation group has been established to develop–through a consensus approach–clinical trial eligibility criteria, case definition, core data collection variables and outcomes. [202]

These criteria, definitions and measures can be integrated into pre-positioned protocols, available for future outbreaks. Pre-positioned protocols have been proposed and adopted for other emerging infections as a way of tackling the challenges of conducting research on sporadic diseases that require a rapid response. [203] Not only do pre-positioned protocols have the advantage of accelerating the pace at which research can start, but they also enable multiple studies to be conducted in a comparable way.

Our systematic review demonstrates the need to collect data on LF clinical characteristics and clinical management in a structured and harmonised fashion with defined core data requirements. This will lead to improvements in the understanding and clinical diagnosis of LF. It will further inform the design of clinical trials of existing and new treatments, and the potential application of indicators for pharmacological treatments, including the severity of disease requiring treatment. Coordinating high-quality research methods across the LF research community can contribute to building a strong evidence base on new treatments for all LF patients and foster confidence in their integration into clinical care.

Search Strategy.

(DOCX)

Click here for additional data file.

Study type definitions.

(DOCX)

Click here for additional data file.

Data dictionary and extraction manual.

(DOCX)

Click here for additional data file.

Other reported signs and symptoms.

(DOCX)

Click here for additional data file.

Prevalent and clinically significant signs and symptoms.

(DOCX)

Click here for additional data file.

Full raw data.

(XLSX)

Click here for additional data file.

29 Jun 2021

Dear Mrs. Bourner,

Thank you very much for submitting your manuscript "Clinical characterization of Lassa fever: a systematic review of clinical reports and research to inform clinical trial design" for consideration at PLOS Neglected Tropical Diseases. As with all papers reviewed by the journal, your manuscript was reviewed by members of the editorial board and by several independent reviewers. In light of the reviews (below this email), we would like to invite the resubmission of a significantly-revised version that takes into account the reviewers' comments.

We cannot make any decision about publication until we have seen the revised manuscript and your response to the reviewers' comments. Your revised manuscript is also likely to be sent to reviewers for further evaluation.

When you are ready to resubmit, please upload the following:

[1] A letter containing a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

[2] Two versions of the revised manuscript: one with either highlights or tracked changes denoting where the text has been changed; the other a clean version (uploaded as the manuscript file).

Important additional instructions are given below your reviewer comments.

Please prepare and submit your revised manuscript within 60 days. If you anticipate any delay, please let us know the expected resubmission date by replying to this email. Please note that revised manuscripts received after the 60-day due date may require evaluation and peer review similar to newly submitted manuscripts.

Thank you again for your submission. We hope that our editorial process has been constructive so far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Manuel Schibler

Associate Editor

PLOS Neglected Tropical Diseases

Andrés Henao-Martínez

Deputy Editor

PLOS Neglected Tropical Diseases

***********************

Reviewer's Responses to Questions

Key Review Criteria Required for Acceptance?

As you describe the new analyses required for acceptance, please consider the following:

Methods

-Are the objectives of the study clearly articulated with a clear testable hypothesis stated?

-Is the study design appropriate to address the stated objectives?

-Is the population clearly described and appropriate for the hypothesis being tested?

-Is the sample size sufficient to ensure adequate power to address the hypothesis being tested?

-Were correct statistical analysis used to support conclusions?

-Are there concerns about ethical or regulatory requirements being met?

Reviewer #1: The objectives of this study was to analyze publication on LF treatment in the effort of finding better standards for LF studies concerning the inclusion/exclusion and outcome criteria as well as the clinical case definition. The study design was very robust concerning the search criteria as well as the analysis of the data. The number of included studies in this analysis is very comprehensive. The statistics are exclusively descriptive, as this study does not formulate a clasical hypothesis. There are no ethical concerns.

Reviewer #2: The study is a systematic review of the literature reporting the clinical signs and symptoms of Lassa fever. The suggested major revisions prior to acceptance are listed below.

1. The link between this review manuscript and how its data might be used to guide clinical trials needs to be better established. In the Introduction section, consider describing how other review papers guided clinical trials for treatments for other febrile illnesses.

2. In the Discussion section consider adding a subsection describing how the results might be used to guide future ribavirin trials. Is there a proposed set of symptoms that should be considered for a suspected case definition?

3. While the focus of the study is to provide information regarding ribavirin trials, there is little information in the manuscript about selected studies that included ribavirin components. How many of the studies included ribavirin? Was ribavirin generally found to be effective in those studies?

4. The premise of this article seems to be that Lassa symptoms can be standardized across geographic regions. What are the implications of standardization? Different countries have different strains of LF. Also, public health units differ by time to clinical presentation and laboratory capacity (for evaluating lab-based items contributing to a suspected case definition). While it may be beyond the scope of this study to report in detail on those challenges, they do seem worth mentioning. Consider adding a subsection in the Discussion section about the challenges of standardization.

5. Consider adding a subsection in the Discussion sections about the study limitations; e.g., test validity, little distinction between acute and long-term exposure. For instance, IgG testing measures long-term exposure and is associated with different symptoms than Ag+ acute cases.

Reviewer #3: (No Response)

--------------------

Results

-Does the analysis presented match the analysis plan?

-Are the results clearly and completely presented?

-Are the figures (Tables, Images) of sufficient quality for clarity?

Reviewer #1: The analysis follows a clear plan, the presentation of the result is clear and comprehensive. All figures and tables are of good quality. It is of note that the results include a broad range of clinical symptoms that go beyond what is often included in clincal case definitions. This corresponds to the numerous case reports that represent the broad clinical picture of LF and is thus an important contribution. The authors also recorded which clinical symptoms were more frequent after baseline.

The authors alos analyzed the frequency of bleeding at different sites and found haematuria the most frequent presentation. However, it should be more thoroughly explained which kinds of haemturia were documented, and if cases of microhaematuria were included, as it can be a resut of kidney injury but does not necessarily represent true bleeding in the stricter sense to my opinion.

Overall, the results are very informative in the effort of finding better standards for clinical studies on LF

Reviewer #2: The analysis does match the analysis plan, the results are clearly and completely presented, and the figures are of sufficient quality for clarity.

Reviewer #3: (No Response)

--------------------

Conclusions

-Are the conclusions supported by the data presented?

-Are the limitations of analysis clearly described?

-Do the authors discuss how these data can be helpful to advance our understanding of the topic under study?

-Is public health relevance addressed?

Reviewer #1: The conclusion are sound in nature but it should be pointed out a little more that this analysis is strictly descariptive and that in order to reach the declared goal of establishing standardized protocols, results would need to be graded (for discussion). the rest of the discussion of the result is very well suited to inform the understanding of the topic.

Reviewer #2: The link between this review manuscript and how its data might be used to guide clinical trials needs to be better established. In the Introduction section, consider describing how other review papers guided clinical trials for treatments for other febrile illnesses.

Reviewer #3: (No Response)

--------------------

Editorial and Data Presentation Modifications?

Use this section for editorial suggestions as well as relatively minor modifications of existing data that would enhance clarity. If the only modifications needed are minor and/or editorial, you may wish to recommend “Minor Revision” or “Accept”.

Reviewer #1: Figure 3 needs to be graphically edited as the numbers are very hard to read, even on a large screen

Other than this minior issue I would recommend to accept

Reviewer #2: Minor suggested revisions:

1. Lines 12-13: "all studies were included"

This statement is overly broad as the literature search did not cover all available bibliographic databases and search engines (e.g., Google Scholar)

2. Line 17: what arose from these discussions?

3. Lines 54-56. Further evidence in the form of clinical trials is required both to confirm the efficacy and safety of ribavirin and to test new therapies.

While this statement is true, consider elaborating on this statement by mentioning the advantages of testing in additional geographic regions, larger sample sizes, accounting for different LF strains, etc.

4. Line 59-60: the clinical characteristics of LF and patient outcomes

The second component of the sentence clinical characteristics of patient outcomes is confusing. Consider rewording.

5. Line 87: spell out acronym for REDCap on first occurrence

6. Lines 102-113: Overuse of “We”

7. Line 225 (Table 5). The definition of laboratory confirmation is extremely influential, particularly at baseline. For instance, Ag ELISA measures acute exposure, IgM measures recent exposure, and IgG measures long-term exposure. It may be helpful to review Table 5 by diagnostic approach to determine whether pooling is justified.

8. Line 262 – “An agreed”

Consider replacing agreed with uniform.

9. Lines 284-287: What are the reasons for this? Limited laboratory capacity?

10. Discussion section: seems to be missing any mention of the need for improved diagnostics and rapid testing

Reviewer #3: (No Response)

--------------------

Summary and General Comments

Use this section to provide overall comments, discuss strengths/weaknesses of the study, novelty, significance, general execution and scholarship. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. If requesting major revision, please articulate the new experiments that are needed.

Reviewer #1: this paper provides a meaningful contribution and is the largest analysis of the kind to my knowledge.

Reviewer #2: General comments

The study addresses a void in the literature by providing a comprehensive systematic review of the literature reporting the clinical signs and symptoms of Lassa fever to guide future ribavirin trials.

Major comments

1. The link between this review manuscript and how its data might be used to guide clinical trials needs to be better established. In the Introduction section, consider describing how other review papers guided clinical trials for treatments for other febrile illnesses.

2. In the Discussion section consider adding a subsection describing how the results might be used to guide future ribavirin trials. Is there a proposed set of symptoms that should be considered for a suspected case definition?

3. While the focus of the study is to provide information regarding ribavirin trials, there is little information in the manuscript about selected studies that included ribavirin components. How many of the studies included ribavirin? Was ribavirin generally found to be effective in those studies?

4. The premise of this article seems to be that Lassa symptoms can be standardized across geographic regions. What are the implications of standardization? Different countries have different strains of LF. Also, public health units differ by time to clinical presentation and laboratory capacity (for evaluating lab-based items contributing to a suspected case definition). While it may be beyond the scope of this study to report in detail on those challenges, they do seem worth mentioning. Consider adding a subsection in the Discussion section about the challenges of standardization.

5. Consider adding a subsection in the Discussion sections about the study limitations; e.g., test validity, little distinction between acute and long-term exposure. For instance, IgG testing measures long-term exposure and is associated with different symptoms than Ag+ acute cases.

Minor comments

1. Lines 12-13: "all studies were included"

This statement is overly broad as the literature search did not cover all available bibliographic databases and search engines (e.g., Google Scholar)

2. Line 17: what arose from these discussions?

3. Lines 54-56. Further evidence in the form of clinical trials is required both to confirm the efficacy and safety of ribavirin and to test new therapies.

While this statement is true, consider elaborating on this statement by mentioning the advantages of testing in additional geographic regions, larger sample sizes, accounting for different LF strains, etc.

4. Line 59-60: the clinical characteristics of LF and patient outcomes

The second component of the sentence clinical characteristics of patient outcomes is confusing. Consider rewording.

5. Line 87: spell out acronym for REDCap on first occurrence

6. Lines 102-113: Overuse of “We”

7. Line 225 (Table 5). The definition of laboratory confirmation is extremely influential, particularly at baseline. For instance, Ag ELISA measures acute exposure, IgM measures recent exposure, and IgG measures long-term exposure. It may be helpful to review Table 5 by diagnostic approach to determine whether pooling is justified.

8. Line 262 – “An agreed”

Consider replacing agreed with uniform.

9. Lines 284-287: What are the reasons for this? Limited laboratory capacity?

10. Discussion section: seems to be missing any mention of the need for improved diagnostics and rapid testing.

Reviewer #3: Merson et al., report valuable information about the signs, symptoms and outcomes of infection with Lassa virus. I believe that the following points are important and deserve further clarification.

More data regarding Lassa virus epidemiology in the 'Introduction' section may further enhance the structural feature of the manuscript. The article " Systematic review and meta-analysis of the epidemiology of Lassa virus in humans, rodents and other mammals in sub-Saharan Africa. PLoS Negl Trop Dis. 2020 Aug 26;14(8):e0008589" may be useful to provide such information to the readers.

Line 61: I would suggest that the authors reword the purpose of the study. “individual-level data” is confusing with the reviews performed on individual patient data from the original databases of the authors of the included studies. This concept of individual data from included study participants should also be clarified throughout the manuscript.

Line 99: The authors should clarify the detection methods and the targets (RNA, viral antigen, antibodies, IgM, IgG…) searched to be considered as confirmed cases of Lassa virus infection.

Line 102-106: In addition to the proportions reported, authors should ideally (if possible) provide the individual data of the included studies.

The signs and symptoms reported in this work should be fully discussed in relation to the multiple other febrile illnesses and haemorrhagic fevers due to other pathogens present in these endemic areas of West Africa and presenting the same clinical picture.

The discussion needs more references to better inform the concept of the research. For example, line 291, what is the work that argues that pregnant women are at higher risk of death from infection with Lassa virus.

The review authors should clearly describe the limitations of this article.

--------------------

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Jeffrey G. Shaffer

Reviewer #3: No

Figure Files:

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

Data Requirements:

Please note that, as a condition of publication, PLOS' data policy requires that you make available all data used to draw the conclusions outlined in your manuscript. Data must be deposited in an appropriate repository, included within the body of the manuscript, or uploaded as supporting information. This includes all numerical values that were used to generate graphs, histograms etc.. For an example see here: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5.

Reproducibility:

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

20 Aug 2021

Submitted filename: Response to Reviewer Comments 20AUG21.docx

Click here for additional data file.

3 Sep 2021

Dear Mrs. Bourner,

We are pleased to inform you that your manuscript 'Clinical characterization of Lassa fever: a systematic review of clinical reports and research to inform clinical trial design' has been provisionally accepted for publication in PLOS Neglected Tropical Diseases.

Before your manuscript can be formally accepted, we would like you to address two more minor concerns:

1. Page 5, line 54: please replace "although this figure is lower (12%) in more recent studies" by "although this figure is lower (12%) in a more recent study"

2. Page 9, lines 150-151: please correct the percentage of quasi-randomised studies; it's 1 (or 1.4), not 2%

You will also need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch with a set of requests.

Please note that your manuscript will not be scheduled for publication until you have made the required changes, so a swift response is appreciated.

IMPORTANT: The editorial review process is now complete. PLOS will only permit corrections to spelling, formatting or significant scientific errors from this point onwards. Requests for major changes, or any which affect the scientific understanding of your work, will cause delays to the publication date of your manuscript.

Should you, your institution's press office or the journal office choose to press release your paper, you will automatically be opted out of early publication. We ask that you notify us now if you or your institution is planning to press release the article. All press must be co-ordinated with PLOS.

Thank you again for supporting Open Access publishing; we are looking forward to publishing your work in PLOS Neglected Tropical Diseases.

Best regards,

Manuel Schibler

Associate Editor

PLOS Neglected Tropical Diseases

Andrés Henao-Martínez

Deputy Editor

PLOS Neglected Tropical Diseases

***********************************************************

Reviewer's Responses to Questions

Key Review Criteria Required for Acceptance?

As you describe the new analyses required for acceptance, please consider the following:

Methods

-Are the objectives of the study clearly articulated with a clear testable hypothesis stated?

-Is the study design appropriate to address the stated objectives?

-Is the population clearly described and appropriate for the hypothesis being tested?

-Is the sample size sufficient to ensure adequate power to address the hypothesis being tested?

-Were correct statistical analysis used to support conclusions?

-Are there concerns about ethical or regulatory requirements being met?

Reviewer #3: (No Response)

**********

Results

-Does the analysis presented match the analysis plan?

-Are the results clearly and completely presented?

-Are the figures (Tables, Images) of sufficient quality for clarity?

Reviewer #3: (No Response)

**********

Conclusions

-Are the conclusions supported by the data presented?

-Are the limitations of analysis clearly described?

-Do the authors discuss how these data can be helpful to advance our understanding of the topic under study?

-Is public health relevance addressed?

Reviewer #3: (No Response)

**********

Editorial and Data Presentation Modifications?

Use this section for editorial suggestions as well as relatively minor modifications of existing data that would enhance clarity. If the only modifications needed are minor and/or editorial, you may wish to recommend “Minor Revision” or “Accept”.

Reviewer #3: (No Response)

**********

Summary and General Comments

Use this section to provide overall comments, discuss strengths/weaknesses of the study, novelty, significance, general execution and scholarship. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. If requesting major revision, please articulate the new experiments that are needed.

Reviewer #3: Although the authors have adequately addressed my suggestions, I think that the citation of the 30% case fatality ratio from the meta-analysis [5] might not be attenuated by the citation [6] which is an original article and therefore difficult to compare. In addition, the authors of the present review report in the abstract the same 30% case fatality ratio found from their own work.

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No

15 Sep 2021

Dear Ms Bourner,

We are delighted to inform you that your manuscript, "Clinical characterization of Lassa fever: a systematic review of clinical reports and research to inform clinical trial design," has been formally accepted for publication in PLOS Neglected Tropical Diseases.

We have now passed your article onto the PLOS Production Department who will complete the rest of the publication process. All authors will receive a confirmation email upon publication.

The corresponding author will soon be receiving a typeset proof for review, to ensure errors have not been introduced during production. Please review the PDF proof of your manuscript carefully, as this is the last chance to correct any scientific or type-setting errors. Please note that major changes, or those which affect the scientific understanding of the work, will likely cause delays to the publication date of your manuscript. Note: Proofs for Front Matter articles (Editorial, Viewpoint, Symposium, Review, etc...) are generated on a different schedule and may not be made available as quickly.

Soon after your final files are uploaded, the early version of your manuscript will be published online unless you opted out of this process. The date of the early version will be your article's publication date. The final article will be published to the same URL, and all versions of the paper will be accessible to readers.

Thank you again for supporting open-access publishing; we are looking forward to publishing your work in PLOS Neglected Tropical Diseases.

Best regards,

Shaden Kamhawi

co-Editor-in-Chief

PLOS Neglected Tropical Diseases

Paul Brindley

co-Editor-in-Chief

PLOS Neglected Tropical Diseases

Table 1

Number of studies included by study type.

Total publications (N)	147
Publication type	N (%)
Full text publication	132 (89.7)
Conference abstract	10 (6.8)
Letter/short communication	5 (3.4)
Study type1	N (%) 2
Case report(s)	53 (36) [32–85]
Case series	41 (28) [86–125]
Cohort study	30 (20) [6,126–154]
Case-control study	10 (7) [155–164]
Cross-sectional study	11 (7) [165–175]
Quasi-randomised	2 (1) [17,176]

1Definitions have been adapted from the Cochrane Community Glossary, available at community.cochrane.org/glossary ()

2 Citations shown in bold italics

Table 2

Individuals and publications included in the systematic review per country.

Country	N individuals (%)	N publications (%)1,2
Nigeria	4461 (52)	61 (41)[6,32,33,37,45,47,54,55,58,68,70,81,84,86–90,93,94,97,99,100,104,109–113,119,121,123,124,126–128,133–135,138,140,141,145,149–152,155–157,159,164,171–173,175]
Sierra Leone	3563 (42)	45 (31)[17,42,46,50,53,67,76,77,82,85,93,95,96,107,118,120,122,130–132,139,142–144,146,148,153,154,158,160–163, 165,169,174,176,177]
Liberia	373 (4)	10 (7)[35,36,69,80,105,106,108,136,148,170]
Benin	73 (1)	2 (1)[39,125]
Guinea	30 (<1)	3 (2)[83,122,129]
United States	16 (<1)	13 (9)[38,43,44,57,60,62,66,73,91,93,103,116,147]
UK	10 (<1)	8 (5)[41,48,49,59,65,78,101,115]
Ghana	6 (<1)	3 (2)[92,98,102]
Germany	4 (<1)	4 (3)[51,61,116,117]
Netherlands	4 (<1)	4 (3)[72,74,75,117]
Canada	2 (<1)	2 (1)[40,64]
Japan	2 (<1)	2 (1)[56,79]
Togo	2 (<1)	1 (1)[114]
Ireland	1 (<1)	1 (1)[63]
Israel	1 (<1)	1 (1)[178]
Ivory Coast	1 (<1)	1 (1)[34]
Sweden	1 (<1)	1 (1)[52]

1N >147 and % >100 as some publications include individuals from >1 country

2Citations shown in bold italics

Table 3

Specified inclusion criteria by type (N = number of studies).

Criteria	N (%)
Publications specifying eligibility criteria (inclusion or exclusion)	68/147 (46)
Publications specifying inclusion criteria	68/68 (100)
Laboratory confirmation of LF either alone or in conjunction with assessment of clinical symptoms	27/68 (40)
Confirmation or diagnosis of LF (method of diagnosis unspecified)	26/68 (38)
Patients with pre-specified signs and symptoms consistent with LF	20/68 (29)
Patients with fever alone or in conjunction with other symptoms	17/68 (25)
Publications specifying exclusion criteria	5/68 (7)
Geographic location	2/5 (40)
Hearing loss (sensorineural or conductive hearing loss)	1/5 (20)
Co-infection with Marburg virus	1/5 (20)
Contacts of index LF case	1/5 (20)
Patients managed on an out-patient basis	1/5 (20)
Patients with an incomplete dataset	1/5 (20)

Table 4

Laboratory Diagnostics—Single and combined testing methods by number of individuals tested.

	Included N (%)*	Exclusively N (%)	RT-PCRN (%)	Viral CultureN (%)	IFAN (%)	IgM ELISA N (%)	Antigen ELISAN (%)	IgG ELISA N (%)	CFN (%)	IHCN (%)	OtherN (%)
RT-PCR	4127 (48)	2348 (27)		1497 (18)	79 (<1)	1549 (18)	268 (3)	156 (2)	0 (0)	1355 (16)	46 (<1)
Viral Culture	3192 (37)	28 (<1)	1497 (18)		1638 (19)	1566 (18)	73 (<1)	213 (2)	118 (1)	1355 (16)	155 (2)
IFA	2232 (26)	99 (1)	79 (<1)	1638 (19)		497 (6)	361 (4)	188 (2)	217 (3)	0 (0)	155 (2)
IgM ELISA	3542 (41)	0 (0)	1549 (18)	1566 (18)	497 (6)		2009 (23)	911 (11)	0 (0)	1354 (16)	642 (8)
Antigen ELISA	2204 (26)	78 (<1)	268 (3)	73 (<1)	361 (4)	2009 (23)		775 (9)	0 (0)	1 (<1)	643 (8)
IgG ELISA	919 (11)	4 (<1)	156 (2)	213 (2)	188 (2)	911 (11)	775 (9)		0 (0)	1 (<1)	643 (8)
CF	307 (4)	30 (<1)	0 (0)	118 (1)	217 (3)	0 (0)	0 (0)	0 (0)		1 (<1)	0 (0)
IHC	1357 (16)	0 (0)	1355 (16)	1355 (16)	0 (0)	1354 (16)	1 (<1)	1 (<1)	1 (<1)		0 (0)
Other**	1187 (14)	389 (5)	46 (<1)	155 (2)	155 (2)	642 (8)	643 (8)	643 (8)	0 (0)	0 (0)

Green = high confidence of acute LF infection; Yellow = moderate confidence; Orange = low confidence.

Abbreviations: Reverse Transcription Polymerase Chain Reaction (RT-PCR); Immunofluoresence Assay (IFA); Immunoglobulin M (IgM); Immunoglobulin G (IgG); Complement Fixation (CF); Immunohistochemistry (IHC)

*Testing method not reported in 3 (2%) publications, including 5 (4%) individuals

** Other includes: Experimental LASV Antigen Rapid Test cassettes and dipstick LFI = 2 (2%) publications and 45 (1%) individuals; Lateral flow immunoassay (LFI) = 2 (2%) publications and 598 (10%) individuals; LF-specific antibody titre = 1 (1) publication and 154 (3%) individuals.

Table 5

Number and percentage of bleeding sites reported.

BLEEDING SITES	Baseline n/N (%)	Post-baseline n/N (%)
(Sub) conjunctiva/eyes	10/168 (6)	12/1088 (1)
Ears	0	1/1088 (<1)
Epistaxis (nose)	8/168 (5)	17/1088 (2)
Gingiva/mouth/buccal mucosa	9/168 (5)	20/1088 (2)
Haematemesis/vomit/upper GI	14/168 (8)	38/1088 (3)
Haematuria/urine	23/168 (14)	131/1088 (12)
Haemoptysis/pleural effusion/cough/respiratory/sputum	1/168 (1)	14/1088 (1)
Needle sites/ wounds	2/168 (1)	13/1088 (1)
Stool/rectum/melena/haematochezia/GI/overt	3/168 (2)	78/1088 (7)
Vagina	1/168 (1)	59/1088 (5)

Table 6

AST ranges in papers reporting AST values.

	Total reporting AST values		All AST values <150 IU/L		All AST values ≥150 IU/L
	Publications N	Individualsn	Publications N (%)	Individualsn (%)	Publications N (%)	Individuals n (%)
AST baseline	29	799	10 (34)	19 (2)	15 (52)	459 (57)
AST anytime	43	1279	9 (21)	18 (1)	14 (33)	284 (22)